Bump version: 0.24.3-beta.0 → 0.24.3

I promise this time I tested it locally :)

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.7.0"
+current_version = "0.21.2"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.
@@ -24,34 +24,51 @@ commit = true
 message = "Bump version: {current_version} → {new_version}"
 commit_args = ""
 
-[tool.bumpversion.parts.pre_l]
-values = ["beta", "final"]
-optional_value = "final"
+# Java maven files
+pre_commit_hooks = [
+  """
+    NEW_VERSION="${BVHOOK_NEW_MAJOR}.${BVHOOK_NEW_MINOR}.${BVHOOK_NEW_PATCH}"
+    if [ ! -z "$BVHOOK_NEW_PRE_L" ] && [ ! -z "$BVHOOK_NEW_PRE_N" ]; then
+        NEW_VERSION="${NEW_VERSION}-${BVHOOK_NEW_PRE_L}.${BVHOOK_NEW_PRE_N}"
+    fi
+    echo "Constructed new version: $NEW_VERSION"
+    cd java && mvn versions:set -DnewVersion=$NEW_VERSION && mvn versions:commit
 
-[[tool.bumpversion.files]]
-filename = "node/package.json"
-search = "\"version\": \"{current_version}\","
-replace = "\"version\": \"{new_version}\","
+    # Check for any modified but unstaged pom.xml files
+    MODIFIED_POMS=$(git ls-files -m | grep pom.xml)
+    if [ ! -z "$MODIFIED_POMS" ]; then
+        echo "The following pom.xml files were modified but not staged. Adding them now:"
+        echo "$MODIFIED_POMS" | while read -r file; do
+            git add "$file"
+            echo "Added: $file"
+        done
+    fi
+    """,
+]
+
+[tool.bumpversion.parts.pre_l]
+optional_value = "final"
+values = ["beta", "final"]
 
 [[tool.bumpversion.files]]
 filename = "nodejs/package.json"
-search = "\"version\": \"{current_version}\","
 replace = "\"version\": \"{new_version}\","
+search = "\"version\": \"{current_version}\","
 
 # nodejs binary packages
 [[tool.bumpversion.files]]
 glob = "nodejs/npm/*/package.json"
-search = "\"version\": \"{current_version}\","
 replace = "\"version\": \"{new_version}\","
+search = "\"version\": \"{current_version}\","
 
 # Cargo files
 # ------------
 [[tool.bumpversion.files]]
-filename = "rust/ffi/node/Cargo.toml"
-search = "\nversion = \"{current_version}\""
+filename = "rust/lancedb/Cargo.toml"
 replace = "\nversion = \"{new_version}\""
+search = "\nversion = \"{current_version}\""
 
 [[tool.bumpversion.files]]
-filename = "rust/lancedb/Cargo.toml"
-search = "\nversion = \"{current_version}\""
+filename = "nodejs/Cargo.toml"
 replace = "\nversion = \"{new_version}\""
+search = "\nversion = \"{current_version}\""

.cargo/config.toml

@@ -31,6 +31,13 @@ rustflags = [
 [target.x86_64-unknown-linux-gnu]
 rustflags = ["-C", "target-cpu=haswell", "-C", "target-feature=+avx2,+fma,+f16c"]
 
+[target.x86_64-unknown-linux-musl]
+rustflags = ["-C", "target-cpu=haswell", "-C", "target-feature=-crt-static,+avx2,+fma,+f16c"]
+
+[target.aarch64-unknown-linux-musl]
+linker = "aarch64-linux-musl-gcc"
+rustflags = ["-C", "target-feature=-crt-static"]
+
 [target.aarch64-apple-darwin]
 rustflags = ["-C", "target-cpu=apple-m1", "-C", "target-feature=+neon,+fp16,+fhm,+dotprod"]
 
@@ -38,3 +45,7 @@ rustflags = ["-C", "target-cpu=apple-m1", "-C", "target-feature=+neon,+fp16,+fhm
 # not found errors on systems that are missing it.
 [target.x86_64-pc-windows-msvc]
 rustflags = ["-Ctarget-feature=+crt-static"]
+
+# Experimental target for Arm64 Windows
+[target.aarch64-pc-windows-msvc]
+rustflags = ["-Ctarget-feature=+crt-static"]

.github/workflows/build_linux_wheel/action.yml

@@ -36,8 +36,7 @@ runs:
         args: ${{ inputs.args }}
         before-script-linux: |
           set -e
-          yum install -y openssl-devel \
-            && curl -L https://github.com/protocolbuffers/protobuf/releases/download/v24.4/protoc-24.4-linux-$(uname -m).zip > /tmp/protoc.zip \
+          curl -L https://github.com/protocolbuffers/protobuf/releases/download/v24.4/protoc-24.4-linux-$(uname -m).zip > /tmp/protoc.zip \
             && unzip /tmp/protoc.zip -d /usr/local \
             && rm /tmp/protoc.zip
     - name: Build Arm Manylinux Wheel
@@ -52,12 +51,7 @@ runs:
         args: ${{ inputs.args }}
         before-script-linux: |
           set -e
-          apt install -y unzip
-          if [ $(uname -m) = "x86_64" ]; then
-            PROTOC_ARCH="x86_64"
-          else
-            PROTOC_ARCH="aarch_64"
-          fi
-          curl -L https://github.com/protocolbuffers/protobuf/releases/download/v24.4/protoc-24.4-linux-$PROTOC_ARCH.zip > /tmp/protoc.zip \
+          yum install -y clang \
+            && curl -L https://github.com/protocolbuffers/protobuf/releases/download/v24.4/protoc-24.4-linux-aarch_64.zip > /tmp/protoc.zip \
             && unzip /tmp/protoc.zip -d /usr/local \
             && rm /tmp/protoc.zip

.github/workflows/build_mac_wheel/action.yml

@@ -20,7 +20,7 @@ runs:
       uses: PyO3/maturin-action@v1
       with:
         command: build
+        # TODO: pass through interpreter
         args: ${{ inputs.args }}
         docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
         working-directory: python
-        interpreter: 3.${{ inputs.python-minor-version }}

.github/workflows/build_windows_wheel/action.yml

@@ -28,7 +28,7 @@ runs:
         args: ${{ inputs.args }}
         docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
         working-directory: python
-    - uses: actions/upload-artifact@v3
+    - uses: actions/upload-artifact@v4
       with:
         name: windows-wheels
         path: python\target\wheels

.github/workflows/cargo-publish.yml

@@ -5,8 +5,8 @@ on:
     tags-ignore:
       # We don't publish pre-releases for Rust. Crates.io is just a source
       # distribution, so we don't need to publish pre-releases.
-      - 'v*-beta*'
-      - '*-v*' # for example, python-vX.Y.Z
+      - "v*-beta*"
+      - "*-v*" # for example, python-vX.Y.Z
 
 env:
   # This env var is used by Swatinem/rust-cache@v2 for the cache
@@ -19,6 +19,8 @@ env:
 jobs:
   build:
     runs-on: ubuntu-22.04
+    permissions:
+      id-token: write
     timeout-minutes: 30
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
@@ -31,6 +33,8 @@ jobs:
         run: |
           sudo apt update
           sudo apt install -y protobuf-compiler libssl-dev
+      - uses: rust-lang/crates-io-auth-action@v1
+        id: auth
       - name: Publish the package
         run: |
-          cargo publish -p lancedb --all-features --token ${{ secrets.CARGO_REGISTRY_TOKEN }}
+          cargo publish -p lancedb --all-features --token ${{ steps.auth.outputs.token }}

.github/workflows/docs.yml

@@ -18,53 +18,49 @@ concurrency:
   group: "pages"
   cancel-in-progress: true
 
+env:
+  # This reduces the disk space needed for the build
+  RUSTFLAGS: "-C debuginfo=0"
+  # according to: https://matklad.github.io/2021/09/04/fast-rust-builds.html
+  # CI builds are faster with incremental disabled.
+  CARGO_INCREMENTAL: "0"
+
 jobs:
   # Single deploy job since we're just deploying
   build:
     environment:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
-    runs-on: buildjet-8vcpu-ubuntu-2204
+    runs-on: ubuntu-24.04
     steps:
       - name: Checkout
         uses: actions/checkout@v4
-      - name: Install dependecies needed for ubuntu
+      - name: Install dependencies needed for ubuntu
         run: |
           sudo apt install -y protobuf-compiler libssl-dev
-          rustup update && rustup default        
+          rustup update && rustup default
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
           python-version: "3.10"
           cache: "pip"
           cache-dependency-path: "docs/requirements.txt"
+      - uses: Swatinem/rust-cache@v2
       - name: Build Python
         working-directory: python
         run: |
-          python -m pip install -e .
-          python -m pip install -r ../docs/requirements.txt
+          python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .
+          python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -r ../docs/requirements.txt
       - name: Set up node
         uses: actions/setup-node@v3
         with:
           node-version: 20
           cache: 'npm'
-          cache-dependency-path: node/package-lock.json
-      - uses: Swatinem/rust-cache@v2
       - name: Install node dependencies
         working-directory: node
         run: |
           sudo apt update
           sudo apt install -y protobuf-compiler libssl-dev
-      - name: Build node
-        working-directory: node
-        run: |
-          npm ci
-          npm run build
-          npm run tsc
-      - name: Create markdown files
-        working-directory: node
-        run: |
-          npx typedoc --plugin typedoc-plugin-markdown --out ../docs/src/javascript src/index.ts
       - name: Build docs
         working-directory: docs
         run: |
@@ -72,9 +68,9 @@ jobs:
       - name: Setup Pages
         uses: actions/configure-pages@v2
       - name: Upload artifact
-        uses: actions/upload-pages-artifact@v1
+        uses: actions/upload-pages-artifact@v3
         with:
           path: "docs/site"
       - name: Deploy to GitHub Pages
         id: deployment
-        uses: actions/deploy-pages@v1
+        uses: actions/deploy-pages@v4

.github/workflows/docs_test.yml

@@ -24,15 +24,19 @@ env:
 jobs:
   test-python:
     name: Test doc python code
-    runs-on: "warp-ubuntu-latest-x64-4x"
+    runs-on: ubuntu-24.04
     steps:
     - name: Checkout
       uses: actions/checkout@v4
     - name: Print CPU capabilities
       run: cat /proc/cpuinfo
+    - name: Install protobuf
+      run: |
+        sudo apt update
+        sudo apt install -y protobuf-compiler
     - name: Install dependecies needed for ubuntu
       run: |
-        sudo apt install -y protobuf-compiler libssl-dev
+        sudo apt install -y libssl-dev
         rustup update && rustup default
     - name: Set up Python
       uses: actions/setup-python@v5
@@ -45,7 +49,7 @@ jobs:
     - name: Build Python
       working-directory: docs/test
       run:
-        python -m pip install -r requirements.txt
+        python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -r requirements.txt
     - name: Create test files
       run: |
         cd docs/test
@@ -54,47 +58,3 @@ jobs:
       run: |
         cd docs/test/python
         for d in *; do cd "$d"; echo "$d".py; python "$d".py; cd ..; done
-  test-node:
-    name: Test doc nodejs code
-    runs-on: "warp-ubuntu-latest-x64-4x"
-    timeout-minutes: 60
-    strategy:
-      fail-fast: false
-    steps:
-    - name: Checkout
-      uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-        lfs: true
-    - name: Print CPU capabilities
-      run: cat /proc/cpuinfo
-    - name: Set up Node
-      uses: actions/setup-node@v4
-      with:
-        node-version: 20
-    - name: Install dependecies needed for ubuntu
-      run: |
-        sudo apt install -y protobuf-compiler libssl-dev
-        rustup update && rustup default
-    - name: Rust cache
-      uses: swatinem/rust-cache@v2
-    - name: Install node dependencies
-      run: |
-        sudo swapoff -a
-        sudo fallocate -l 8G /swapfile
-        sudo chmod 600 /swapfile
-        sudo mkswap /swapfile
-        sudo swapon /swapfile
-        sudo swapon --show
-        cd node
-        npm ci
-        npm run build-release
-        cd ../docs
-        npm install
-    - name: Test
-      env:
-        LANCEDB_URI: ${{ secrets.LANCEDB_URI }}
-        LANCEDB_DEV_API_KEY: ${{ secrets.LANCEDB_DEV_API_KEY }}
-      run: |
-        cd docs
-        npm t

.github/workflows/java-publish.yml

@@ -0,0 +1,114 @@
+name: Build and publish Java packages
+on:
+  release:
+    types: [released]
+  pull_request:
+    paths:
+      - .github/workflows/java-publish.yml
+
+jobs:
+  macos-arm64:
+    name: Build on MacOS Arm64
+    runs-on: macos-14
+    timeout-minutes: 45
+    defaults:
+      run:
+        working-directory: ./java/core/lancedb-jni
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: Swatinem/rust-cache@v2
+      - name: Install dependencies
+        run: |
+          brew install protobuf
+      - name: Build release
+        run: |
+          cargo build --release
+      - uses: actions/upload-artifact@v4
+        with:
+          name: liblancedb_jni_darwin_aarch64.zip
+          path: target/release/liblancedb_jni.dylib
+          retention-days: 1
+          if-no-files-found: error
+  linux-arm64:
+    name: Build on Linux Arm64
+    runs-on: warp-ubuntu-2204-arm64-8x
+    timeout-minutes: 45
+    defaults:
+      run:
+        working-directory: ./java/core/lancedb-jni
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: Swatinem/rust-cache@v2
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          toolchain: "1.81.0"
+          cache-workspaces: "./java/core/lancedb-jni"
+          # Disable full debug symbol generation to speed up CI build and keep memory down
+          # "1" means line tables only, which is useful for panic tracebacks.
+          rustflags: "-C debuginfo=1"
+      - name: Install dependencies
+        run: |
+          sudo apt -y -qq update
+          sudo apt install -y protobuf-compiler libssl-dev pkg-config
+      - name: Build release
+        run: |
+          cargo build --release
+      - uses: actions/upload-artifact@v4
+        with:
+          name: liblancedb_jni_linux_aarch64.zip
+          path: target/release/liblancedb_jni.so
+          retention-days: 1
+          if-no-files-found: error
+  linux-x86:
+    runs-on: warp-ubuntu-2204-x64-8x
+    timeout-minutes: 30
+    needs: [macos-arm64, linux-arm64]
+    defaults:
+      run:
+        working-directory: ./java
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: Swatinem/rust-cache@v2
+      - name: Set up Java 8
+        uses: actions/setup-java@v4
+        with:
+          distribution: temurin
+          java-version: 8
+          cache: "maven"
+          server-id: ossrh
+          server-username: SONATYPE_USER
+          server-password: SONATYPE_TOKEN
+          gpg-private-key: ${{ secrets.GPG_PRIVATE_KEY }}
+          gpg-passphrase: ${{ secrets.GPG_PASSPHRASE }}
+      - name: Install dependencies
+        run: |
+          sudo apt -y -qq update
+          sudo apt install -y protobuf-compiler libssl-dev pkg-config
+      - name: Download artifact
+        uses: actions/download-artifact@v4
+      - name: Copy native libs
+        run: |
+          mkdir -p ./core/target/classes/nativelib/darwin-aarch64 ./core/target/classes/nativelib/linux-aarch64
+          cp ../liblancedb_jni_darwin_aarch64.zip/liblancedb_jni.dylib ./core/target/classes/nativelib/darwin-aarch64/liblancedb_jni.dylib
+          cp ../liblancedb_jni_linux_aarch64.zip/liblancedb_jni.so ./core/target/classes/nativelib/linux-aarch64/liblancedb_jni.so
+      - name: Dry run
+        if: github.event_name == 'pull_request'
+        run: |
+          mvn --batch-mode -DskipTests -Drust.release.build=true package
+      - name: Set github
+        run: |
+          git config --global user.email "LanceDB Github Runner"
+          git config --global user.name "dev+gha@lancedb.com"
+      - name: Publish with Java 8
+        if: github.event_name == 'release'
+        run: |
+          echo "use-agent" >> ~/.gnupg/gpg.conf
+          echo "pinentry-mode loopback" >> ~/.gnupg/gpg.conf
+          export GPG_TTY=$(tty)
+          mvn --batch-mode -DskipTests -Drust.release.build=true -DpushChanges=false -Dgpg.passphrase=${{ secrets.GPG_PASSPHRASE }} deploy -P deploy-to-ossrh
+        env:
+          SONATYPE_USER: ${{ secrets.SONATYPE_USER }}
+          SONATYPE_TOKEN: ${{ secrets.SONATYPE_TOKEN }}

.github/workflows/java.yml

@@ -3,6 +3,8 @@ on:
   push:
     branches:
       - main
+    paths:
+      - java/**
   pull_request:
     paths:
       - java/**
@@ -21,9 +23,9 @@ env:
   CARGO_INCREMENTAL: "0"
   CARGO_BUILD_JOBS: "1"
 jobs:
-  linux-build:
+  linux-build-java-11:
     runs-on: ubuntu-22.04
-    name: ubuntu-22.04 + Java 11 & 17
+    name: ubuntu-22.04 + Java 11
     defaults:
       run:
         working-directory: ./java
@@ -33,6 +35,9 @@ jobs:
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: java/core/lancedb-jni
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          components: rustfmt
       - name: Run cargo fmt
         run: cargo fmt --check
         working-directory: ./java/core/lancedb-jni
@@ -40,13 +45,6 @@ jobs:
         run: |
           sudo apt update
           sudo apt install -y protobuf-compiler libssl-dev
-      - name: Install Java 17
-        uses: actions/setup-java@v4
-        with:
-          distribution: temurin
-          java-version: 17
-          cache: "maven"
-      - run: echo "JAVA_17=$JAVA_HOME" >> $GITHUB_ENV
       - name: Install Java 11
         uses: actions/setup-java@v4
         with:
@@ -61,6 +59,41 @@ jobs:
       #   run: cargo clippy --all-targets -- -D warnings
       - name: Running tests with Java 11
         run: mvn clean test
+  linux-build-java-17:
+    runs-on: ubuntu-22.04
+    name: ubuntu-22.04 + Java 17
+    defaults:
+      run:
+        working-directory: ./java
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: java/core/lancedb-jni
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          components: rustfmt
+      - name: Run cargo fmt
+        run: cargo fmt --check
+        working-directory: ./java/core/lancedb-jni
+      - name: Install dependencies
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler libssl-dev
+      - name: Install Java 17
+        uses: actions/setup-java@v4
+        with:
+          distribution: temurin
+          java-version: 17
+          cache: "maven"
+      - run: echo "JAVA_17=$JAVA_HOME" >> $GITHUB_ENV
+      - name: Java Style Check
+        run: mvn checkstyle:check
+      # Disable because of issues in lancedb rust core code
+      # - name: Rust Clippy
+      #   working-directory: java/core/lancedb-jni
+      #   run: cargo clippy --all-targets -- -D warnings
       - name: Running tests with Java 17
         run: |
           export JAVA_TOOL_OPTIONS="$JAVA_TOOL_OPTIONS \

.github/workflows/license-header-check.yml

@@ -0,0 +1,31 @@
+name: Check license headers
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    paths:
+      - rust/**
+      - python/**
+      - nodejs/**
+      - java/**
+      - .github/workflows/license-header-check.yml
+jobs:
+  check-licenses:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v4
+      - name: Install license-header-checker
+        working-directory: /tmp
+        run: |
+            curl -s https://raw.githubusercontent.com/lluissm/license-header-checker/master/install.sh | bash
+            mv /tmp/bin/license-header-checker /usr/local/bin/
+      - name: Check license headers (rust)
+        run: license-header-checker -a -v ./rust/license_header.txt ./ rs && [[ -z `git status -s` ]]
+      - name: Check license headers (python)
+        run: license-header-checker -a -v ./python/license_header.txt python py && [[ -z `git status -s` ]]
+      - name: Check license headers (typescript)
+        run: license-header-checker -a -v ./nodejs/license_header.txt nodejs ts && [[ -z `git status -s` ]]
+      - name: Check license headers (java)
+        run: license-header-checker -a -v ./nodejs/license_header.txt java java && [[ -z `git status -s` ]]

.github/workflows/make-release-commit.yml

@@ -30,7 +30,7 @@ on:
         default: true
         type: boolean
       other:
-        description: 'Make a Node/Rust release'
+        description: 'Make a Node/Rust/Java release'
         required: true
         default: true
         type: boolean
@@ -43,7 +43,7 @@ on:
 jobs:
   make-release:
     # Creates tag and GH release. The GH release will trigger the build and release jobs.
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-24.04
     permissions:
       contents: write
     steps:
@@ -57,15 +57,14 @@ jobs:
           # trigger any workflows watching for new tags. See:
           # https://docs.github.com/en/actions/using-workflows/triggering-a-workflow#triggering-a-workflow-from-a-workflow
           token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+      - name: Validate Lance dependency is at stable version
+        if: ${{ inputs.type == 'stable' }}
+        run: python ci/validate_stable_lance.py
       - name: Set git configs for bumpversion
         shell: bash
         run: |
           git config user.name 'Lance Release'
           git config user.email 'lance-dev@lancedb.com'
-      - name: Set up Python 3.11
-        uses: actions/setup-python@v5
-        with:
-          python-version: "3.11"
       - name: Bump Python version
         if: ${{ inputs.python }}
         working-directory: python
@@ -85,6 +84,7 @@ jobs:
         run: |
           pip install bump-my-version PyGithub packaging
           bash ci/bump_version.sh ${{ inputs.type }} ${{ inputs.bump-minor }} v $COMMIT_BEFORE_BUMP
+          bash ci/update_lockfiles.sh --amend
       - name: Push new version tag
         if: ${{ !inputs.dry_run }}
         uses: ad-m/github-push-action@master
@@ -93,7 +93,3 @@ jobs:
           github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
           branch: ${{ github.ref }}
           tags: true
-      - uses: ./.github/workflows/update_package_lock
-        if: ${{ !inputs.dry_run && inputs.other }}
-        with:
-          github_token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/node.yml

@@ -1,147 +0,0 @@
-name: Node
-
-on:
-  push:
-    branches:
-      - main
-  pull_request:
-    paths:
-      - node/**
-      - rust/ffi/node/**
-      - .github/workflows/node.yml
-      - docker-compose.yml
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-  cancel-in-progress: true
-
-env:
-  # Disable full debug symbol generation to speed up CI build and keep memory down
-  # "1" means line tables only, which is useful for panic tracebacks.
-  #
-  # Use native CPU to accelerate tests if possible, especially for f16
-  # target-cpu=haswell fixes failing ci build
-  RUSTFLAGS: "-C debuginfo=1 -C target-cpu=haswell -C target-feature=+f16c,+avx2,+fma"
-  RUST_BACKTRACE: "1"
-
-jobs:
-  linux:
-    name: Linux (Node ${{ matrix.node-version }})
-    timeout-minutes: 30
-    strategy:
-      matrix:
-        node-version: [ "18", "20" ]
-    runs-on: "ubuntu-22.04"
-    defaults:
-      run:
-        shell: bash
-        working-directory: node
-    steps:
-    - uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-        lfs: true
-    - uses: actions/setup-node@v3
-      with:
-        node-version: ${{ matrix.node-version }}
-        cache: 'npm'
-        cache-dependency-path: node/package-lock.json
-    - uses: Swatinem/rust-cache@v2
-    - name: Install dependencies
-      run: |
-        sudo apt update
-        sudo apt install -y protobuf-compiler libssl-dev
-    - name: Build
-      run: |
-        npm ci
-        npm run build
-        npm run pack-build
-        npm install --no-save ./dist/lancedb-vectordb-*.tgz
-        # Remove index.node to test with dependency installed
-        rm index.node
-    - name: Test
-      run: npm run test
-  macos:
-    timeout-minutes: 30
-    runs-on: "macos-13"
-    defaults:
-      run:
-        shell: bash
-        working-directory: node
-    steps:
-    - uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-        lfs: true
-    - uses: actions/setup-node@v3
-      with:
-        node-version: 20
-        cache: 'npm'
-        cache-dependency-path: node/package-lock.json
-    - uses: Swatinem/rust-cache@v2
-    - name: Install dependencies
-      run: brew install protobuf
-    - name: Build
-      run: |
-        npm ci
-        npm run build
-        npm run pack-build
-        npm install --no-save ./dist/lancedb-vectordb-*.tgz
-        # Remove index.node to test with dependency installed
-        rm index.node
-    - name: Test
-      run: |
-        npm run test
-  aws-integtest:
-    timeout-minutes: 45
-    runs-on: "ubuntu-22.04"
-    defaults:
-      run:
-        shell: bash
-        working-directory: node
-    env:
-      AWS_ACCESS_KEY_ID: ACCESSKEY
-      AWS_SECRET_ACCESS_KEY: SECRETKEY
-      AWS_DEFAULT_REGION: us-west-2
-      # this one is for s3
-      AWS_ENDPOINT: http://localhost:4566
-      # this one is for dynamodb
-      DYNAMODB_ENDPOINT: http://localhost:4566
-      ALLOW_HTTP: true
-    steps:
-    - uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-        lfs: true
-    - uses: actions/setup-node@v3
-      with:
-        node-version: 20
-        cache: 'npm'
-        cache-dependency-path: node/package-lock.json
-    - name: start local stack
-      run: docker compose -f ../docker-compose.yml up -d --wait
-    - name: create s3
-      run: aws s3 mb s3://lancedb-integtest --endpoint $AWS_ENDPOINT
-    - name: create ddb
-      run: |
-        aws dynamodb create-table \
-          --table-name lancedb-integtest \
-          --attribute-definitions '[{"AttributeName": "base_uri", "AttributeType": "S"}, {"AttributeName": "version", "AttributeType": "N"}]' \
-          --key-schema '[{"AttributeName": "base_uri", "KeyType": "HASH"}, {"AttributeName": "version", "KeyType": "RANGE"}]' \
-          --provisioned-throughput '{"ReadCapacityUnits": 10, "WriteCapacityUnits": 10}' \
-          --endpoint-url $DYNAMODB_ENDPOINT
-    - uses: Swatinem/rust-cache@v2
-    - name: Install dependencies
-      run: |
-        sudo apt update
-        sudo apt install -y protobuf-compiler libssl-dev
-    - name: Build
-      run: |
-        npm ci
-        npm run build
-        npm run pack-build
-        npm install --no-save ./dist/lancedb-vectordb-*.tgz
-        # Remove index.node to test with dependency installed
-        rm index.node
-    - name: Test
-      run: npm run integration-test

.github/workflows/nodejs.yml

@@ -47,12 +47,18 @@ jobs:
       run: |
         sudo apt update
         sudo apt install -y protobuf-compiler libssl-dev
+    - uses: actions-rust-lang/setup-rust-toolchain@v1
+      with:
+        components: rustfmt, clippy
     - name: Lint
       run: |
         cargo fmt --all -- --check
         cargo clippy --all --all-features -- -D warnings
         npm ci
         npm run lint-ci
+    - name: Lint examples
+      working-directory: nodejs/examples
+      run: npm ci && npm run lint-ci
   linux:
     name: Linux (NodeJS ${{ matrix.node-version }})
     timeout-minutes: 30
@@ -73,7 +79,7 @@ jobs:
       with:
         node-version: ${{ matrix.node-version }}
         cache: 'npm'
-        cache-dependency-path: node/package-lock.json
+        cache-dependency-path: nodejs/package-lock.json
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |
@@ -91,6 +97,30 @@ jobs:
       env:
         S3_TEST: "1"
       run: npm run test
+    - name: Setup examples
+      working-directory: nodejs/examples
+      run: npm ci
+    - name: Test examples
+      working-directory: ./
+      env:
+        OPENAI_API_KEY: test
+        OPENAI_BASE_URL: http://0.0.0.0:8000
+      run: |
+        python ci/mock_openai.py &
+        cd nodejs/examples
+        npm test
+    - name: Check docs
+      run: |
+        # We run this as part of the job because the binary needs to be built
+        # first to export the types of the native code.
+        set -e
+        npm ci
+        npm run docs
+        if ! git diff --exit-code -- . ':(exclude)Cargo.lock'; then
+          echo "Docs need to be updated"
+          echo "Run 'npm run docs', fix any warnings, and commit the changes."
+          exit 1
+        fi
   macos:
     timeout-minutes: 30
     runs-on: "macos-14"
@@ -107,7 +137,7 @@ jobs:
       with:
         node-version: 20
         cache: 'npm'
-        cache-dependency-path: node/package-lock.json
+        cache-dependency-path: nodejs/package-lock.json
     - uses: Swatinem/rust-cache@v2
     - name: Install dependencies
       run: |

.github/workflows/npm-publish.yml

@@ -1,374 +1,32 @@
 name: NPM Publish
 
+env:
+  MACOSX_DEPLOYMENT_TARGET: '10.13'
+  CARGO_INCREMENTAL: '0'
+
+permissions:
+  contents: write
+  id-token: write
+
 on:
   push:
+    branches:
+      - main
     tags:
       - "v*"
+  pull_request:
+    # This should trigger a dry run (we skip the final publish step)
+    paths:
+      - .github/workflows/npm-publish.yml
+      - Cargo.toml # Change in dependency frequently breaks builds
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
 
 jobs:
-  node:
-    runs-on: ubuntu-latest
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    defaults:
-      run:
-        shell: bash
-        working-directory: node
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - uses: actions/setup-node@v3
-        with:
-          node-version: 20
-          cache: "npm"
-          cache-dependency-path: node/package-lock.json
-      - name: Install dependencies
-        run: |
-          sudo apt update
-          sudo apt install -y protobuf-compiler libssl-dev
-      - name: Build
-        run: |
-          npm ci
-          npm run tsc
-          npm pack
-      - name: Upload Linux Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: node-package
-          path: |
-            node/vectordb-*.tgz
-
-  node-macos:
-    strategy:
-      matrix:
-        config:
-          - arch: x86_64-apple-darwin
-            runner: macos-13
-          - arch: aarch64-apple-darwin
-            # xlarge is implicitly arm64.
-            runner: macos-14
-    runs-on: ${{ matrix.config.runner }}
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install system dependencies
-        run: brew install protobuf
-      - name: Install npm dependencies
-        run: |
-          cd node
-          npm ci
-      - name: Build MacOS native node modules
-        run: bash ci/build_macos_artifacts.sh ${{ matrix.config.arch }}
-      - name: Upload Darwin Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: node-native-darwin-${{ matrix.config.arch }}
-          path: |
-            node/dist/lancedb-vectordb-darwin*.tgz
-
-  nodejs-macos:
-    strategy:
-      matrix:
-        config:
-          - arch: x86_64-apple-darwin
-            runner: macos-13
-          - arch: aarch64-apple-darwin
-            # xlarge is implicitly arm64.
-            runner: macos-14
-    runs-on: ${{ matrix.config.runner }}
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install system dependencies
-        run: brew install protobuf
-      - name: Install npm dependencies
-        run: |
-          cd nodejs
-          npm ci
-      - name: Build MacOS native nodejs modules
-        run: bash ci/build_macos_artifacts_nodejs.sh ${{ matrix.config.arch }}
-      - name: Upload Darwin Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: nodejs-native-darwin-${{ matrix.config.arch }}
-          path: |
-            nodejs/dist/*.node
-
-  node-linux:
-    name: node-linux (${{ matrix.config.arch}}-unknown-linux-gnu
-    runs-on: ${{ matrix.config.runner }}
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    strategy:
-      fail-fast: false
-      matrix:
-        config:
-          - arch: x86_64
-            runner: ubuntu-latest
-          - arch: aarch64
-            # For successful fat LTO builds, we need a large runner to avoid OOM errors.
-            runner: warp-ubuntu-latest-arm64-4x
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      # To avoid OOM errors on ARM, we create a swap file.
-      - name: Configure aarch64 build
-        if: ${{ matrix.config.arch == 'aarch64' }}
-        run: |
-          free -h
-          sudo fallocate -l 16G /swapfile
-          sudo chmod 600 /swapfile
-          sudo mkswap /swapfile
-          sudo swapon /swapfile
-          echo "/swapfile swap swap defaults 0 0" >> sudo /etc/fstab
-          # print info
-          swapon --show
-          free -h
-      - name: Build Linux Artifacts
-        run: |
-          bash ci/build_linux_artifacts.sh ${{ matrix.config.arch }}
-      - name: Upload Linux Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: node-native-linux-${{ matrix.config.arch }}
-          path: |
-            node/dist/lancedb-vectordb-linux*.tgz
-
-  nodejs-linux:
-    name: nodejs-linux (${{ matrix.config.arch}}-unknown-linux-gnu
-    runs-on: ${{ matrix.config.runner }}
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    strategy:
-      fail-fast: false
-      matrix:
-        config:
-          - arch: x86_64
-            runner: ubuntu-latest
-          - arch: aarch64
-            # For successful fat LTO builds, we need a large runner to avoid OOM errors.
-            runner: buildjet-16vcpu-ubuntu-2204-arm
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      # Buildjet aarch64 runners have only 1.5 GB RAM per core, vs 3.5 GB per core for
-      # x86_64 runners. To avoid OOM errors on ARM, we create a swap file.
-      - name: Configure aarch64 build
-        if: ${{ matrix.config.arch == 'aarch64' }}
-        run: |
-          free -h
-          sudo fallocate -l 16G /swapfile
-          sudo chmod 600 /swapfile
-          sudo mkswap /swapfile
-          sudo swapon /swapfile
-          echo "/swapfile swap swap defaults 0 0" >> sudo /etc/fstab
-          # print info
-          swapon --show
-          free -h
-      - name: Build Linux Artifacts
-        run: |
-          bash ci/build_linux_artifacts_nodejs.sh ${{ matrix.config.arch }}
-      - name: Upload Linux Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: nodejs-native-linux-${{ matrix.config.arch }}
-          path: |
-            nodejs/dist/*.node
-      # The generic files are the same in all distros so we just pick
-      # one to do the upload.
-      - name: Upload Generic Artifacts
-        if: ${{ matrix.config.arch == 'x86_64' }}
-        uses: actions/upload-artifact@v4
-        with:
-          name: nodejs-dist
-          path: |
-            nodejs/dist/*
-            !nodejs/dist/*.node
-
-  node-windows:
-    runs-on: windows-2022
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    strategy:
-      fail-fast: false
-      matrix:
-        target: [x86_64-pc-windows-msvc]
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Protoc v21.12
-        working-directory: C:\
-        run: |
-          New-Item -Path 'C:\protoc' -ItemType Directory
-          Set-Location C:\protoc
-          Invoke-WebRequest https://github.com/protocolbuffers/protobuf/releases/download/v21.12/protoc-21.12-win64.zip -OutFile C:\protoc\protoc.zip
-          7z x protoc.zip
-          Add-Content $env:GITHUB_PATH "C:\protoc\bin"
-        shell: powershell
-      - name: Install npm dependencies
-        run: |
-          cd node
-          npm ci
-      - name: Build Windows native node modules
-        run: .\ci\build_windows_artifacts.ps1 ${{ matrix.target }}
-      - name: Upload Windows Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: node-native-windows
-          path: |
-            node/dist/lancedb-vectordb-win32*.tgz
-
-  nodejs-windows:
-    runs-on: windows-2022
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    strategy:
-      fail-fast: false
-      matrix:
-        target: [x86_64-pc-windows-msvc]
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Install Protoc v21.12
-        working-directory: C:\
-        run: |
-          New-Item -Path 'C:\protoc' -ItemType Directory
-          Set-Location C:\protoc
-          Invoke-WebRequest https://github.com/protocolbuffers/protobuf/releases/download/v21.12/protoc-21.12-win64.zip -OutFile C:\protoc\protoc.zip
-          7z x protoc.zip
-          Add-Content $env:GITHUB_PATH "C:\protoc\bin"
-        shell: powershell
-      - name: Install npm dependencies
-        run: |
-          cd nodejs
-          npm ci
-      - name: Build Windows native node modules
-        run: .\ci\build_windows_artifacts_nodejs.ps1 ${{ matrix.target }}
-      - name: Upload Windows Artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: nodejs-native-windows
-          path: |
-            nodejs/dist/*.node
-
-  release:
-    needs: [node, node-macos, node-linux, node-windows]
-    runs-on: ubuntu-latest
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    steps:
-      - uses: actions/download-artifact@v4
-        with:
-          pattern: node-*
-      - name: Display structure of downloaded files
-        run: ls -R
-      - uses: actions/setup-node@v3
-        with:
-          node-version: 20
-          registry-url: "https://registry.npmjs.org"
-      - name: Publish to NPM
-        env:
-          NODE_AUTH_TOKEN: ${{ secrets.LANCEDB_NPM_REGISTRY_TOKEN }}
-        run: |
-          # Tag beta as "preview" instead of default "latest". See lancedb 
-          # npm publish step for more info.
-          if [[ $GITHUB_REF =~ refs/tags/v(.*)-beta.* ]]; then
-            PUBLISH_ARGS="--tag preview"
-          fi
-
-          mv */*.tgz .
-          for filename in *.tgz; do
-            npm publish $PUBLISH_ARGS $filename
-          done
-
-  release-nodejs:
-    needs: [nodejs-macos, nodejs-linux, nodejs-windows]
-    runs-on: ubuntu-latest
-    # Only runs on tags that matches the make-release action
-    if: startsWith(github.ref, 'refs/tags/v')
-    defaults:
-      run:
-        shell: bash
-        working-directory: nodejs
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - uses: actions/download-artifact@v4
-        with:
-          name: nodejs-dist
-          path: nodejs/dist
-      - uses: actions/download-artifact@v4
-        name: Download arch-specific binaries
-        with:
-          pattern: nodejs-*
-          path: nodejs/nodejs-artifacts
-          merge-multiple: true
-      - name: Display structure of downloaded files
-        run: find .
-      - uses: actions/setup-node@v3
-        with:
-          node-version: 20
-          registry-url: "https://registry.npmjs.org"
-      - name: Install napi-rs
-        run: npm install -g @napi-rs/cli
-      - name: Prepare artifacts
-        run: npx napi artifacts -d nodejs-artifacts
-      - name: Display structure of staged files
-        run: find npm
-      - name: Publish to NPM
-        env:
-          NODE_AUTH_TOKEN: ${{ secrets.LANCEDB_NPM_REGISTRY_TOKEN }}
-        # By default, things are published to the latest tag. This is what is
-        # installed by default if the user does not specify a version. This is
-        # good for stable releases, but for pre-releases, we want to publish to
-        # the "preview" tag so they can install with `npm install lancedb@preview`.
-        # See: https://medium.com/@mbostock/prereleases-and-npm-e778fc5e2420
-        run: |
-          if [[ $GITHUB_REF =~ refs/tags/v(.*)-beta.* ]]; then
-            npm publish --access public --tag preview
-          else
-            npm publish --access public
-          fi
-
-  update-package-lock:
-    needs: [release]
-    runs-on: ubuntu-latest
-    permissions:
-      contents: write
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-        with:
-          ref: main
-          persist-credentials: false
-          fetch-depth: 0
-          lfs: true
-      - uses: ./.github/workflows/update_package_lock
-        with:
-          github_token: ${{ secrets.GITHUB_TOKEN }}
-
-  update-package-lock-nodejs:
-    needs: [release-nodejs]
-    runs-on: ubuntu-latest
-    permissions:
-      contents: write
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-        with:
-          ref: main
-          persist-credentials: false
-          fetch-depth: 0
-          lfs: true
-      - uses: ./.github/workflows/update_package_lock_nodejs
-        with:
-          github_token: ${{ secrets.GITHUB_TOKEN }}
-
   gh-release:
+    if: startsWith(github.ref, 'refs/tags/v')
     runs-on: ubuntu-latest
     permissions:
       contents: write
@@ -433,3 +91,277 @@ jobs:
           generate_release_notes: false
           name: Node/Rust LanceDB v${{ steps.extract_version.outputs.version }}
           body: ${{ steps.release_notes.outputs.changelog }}
+
+  build-lancedb:
+    strategy:
+      fail-fast: false
+      matrix:
+        settings:
+          - target: x86_64-apple-darwin
+            host: macos-latest
+            features: ","
+            pre_build: |-
+                brew install protobuf
+                rustup target add x86_64-apple-darwin
+          - target: aarch64-apple-darwin
+            host: macos-latest
+            features: fp16kernels
+            pre_build: brew install protobuf
+          - target: x86_64-pc-windows-msvc
+            host: windows-latest
+            features: ","
+            pre_build: |-
+              choco install --no-progress protoc ninja nasm
+              tail -n 1000 /c/ProgramData/chocolatey/logs/chocolatey.log
+              # There is an issue where choco doesn't add nasm to the path
+              export PATH="$PATH:/c/Program Files/NASM"
+              nasm -v
+          - target: aarch64-pc-windows-msvc
+            host: windows-latest
+            features: ","
+            pre_build: |-
+                choco install --no-progress protoc
+                rustup target add aarch64-pc-windows-msvc
+          - target: x86_64-unknown-linux-gnu
+            host: ubuntu-latest
+            features: fp16kernels
+            # https://github.com/napi-rs/napi-rs/blob/main/debian.Dockerfile
+            docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-debian
+            pre_build: |-
+              set -e &&
+              apt-get update &&
+              apt-get install -y protobuf-compiler pkg-config
+          - target: x86_64-unknown-linux-musl
+            # This one seems to need some extra memory
+            host: ubuntu-2404-8x-x64
+            # https://github.com/napi-rs/napi-rs/blob/main/alpine.Dockerfile
+            docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-alpine
+            features: fp16kernels
+            pre_build: |-
+              set -e &&
+              apk add protobuf-dev curl &&
+              ln -s /usr/lib/gcc/x86_64-alpine-linux-musl/14.2.0/crtbeginS.o /usr/lib/crtbeginS.o &&
+              ln -s /usr/lib/libgcc_s.so /usr/lib/libgcc.so &&
+              CC=gcc &&
+              CXX=g++
+          - target: aarch64-unknown-linux-gnu
+            host: ubuntu-2404-8x-x64
+            # https://github.com/napi-rs/napi-rs/blob/main/debian-aarch64.Dockerfile
+            docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-debian-aarch64
+            features: "fp16kernels"
+            pre_build: |-
+              set -e &&
+              apt-get update &&
+              apt-get install -y protobuf-compiler pkg-config &&
+              # https://github.com/aws/aws-lc-rs/issues/737#issuecomment-2725918627
+              ln -s /usr/aarch64-unknown-linux-gnu/lib/gcc/aarch64-unknown-linux-gnu/4.8.5/crtbeginS.o /usr/aarch64-unknown-linux-gnu/aarch64-unknown-linux-gnu/sysroot/usr/lib/crtbeginS.o &&
+              ln -s /usr/aarch64-unknown-linux-gnu/lib/gcc /usr/aarch64-unknown-linux-gnu/aarch64-unknown-linux-gnu/sysroot/usr/lib/gcc &&
+              rustup target add aarch64-unknown-linux-gnu
+          - target: aarch64-unknown-linux-musl
+            host: ubuntu-2404-8x-x64
+            # https://github.com/napi-rs/napi-rs/blob/main/alpine.Dockerfile
+            docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-alpine
+            features: ","
+            pre_build: |-
+              set -e &&
+              apk add protobuf-dev &&
+              rustup target add aarch64-unknown-linux-musl &&
+              export CC_aarch64_unknown_linux_musl=aarch64-linux-musl-gcc &&
+              export CXX_aarch64_unknown_linux_musl=aarch64-linux-musl-g++
+    name: build - ${{ matrix.settings.target }}
+    runs-on: ${{ matrix.settings.host }}
+    defaults:
+      run:
+        working-directory: nodejs
+    steps:
+      - uses: actions/checkout@v4
+      - name: Setup node
+        uses: actions/setup-node@v4
+        if: ${{ !matrix.settings.docker }}
+        with:
+          node-version: 20
+          cache: npm
+          cache-dependency-path: nodejs/package-lock.json
+      - name: Install
+        uses: dtolnay/rust-toolchain@stable
+        if: ${{ !matrix.settings.docker }}
+        with:
+          toolchain: stable
+          targets: ${{ matrix.settings.target }}
+      - name: Cache cargo
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/registry/index/
+            ~/.cargo/registry/cache/
+            ~/.cargo/git/db/
+            .cargo-cache
+            target/
+          key: nodejs-${{ matrix.settings.target }}-cargo-${{ matrix.settings.host }}
+      - name: Setup toolchain
+        run: ${{ matrix.settings.setup }}
+        if: ${{ matrix.settings.setup }}
+        shell: bash
+      - name: Install dependencies
+        run: npm ci
+      - name: Build in docker
+        uses: addnab/docker-run-action@v3
+        if: ${{ matrix.settings.docker }}
+        with:
+          image: ${{ matrix.settings.docker }}
+          options: "--user 0:0 -v ${{ github.workspace }}/.cargo-cache/git/db:/usr/local/cargo/git/db \
+                    -v ${{ github.workspace }}/.cargo/registry/cache:/usr/local/cargo/registry/cache \
+                    -v ${{ github.workspace }}/.cargo/registry/index:/usr/local/cargo/registry/index \
+                    -v ${{ github.workspace }}:/build -w /build/nodejs"
+          run: |
+            set -e
+            ${{ matrix.settings.pre_build }}
+            npx napi build --platform  --release --no-const-enum \
+              --features ${{ matrix.settings.features }} \
+              --target ${{ matrix.settings.target }} \
+              --dts ../lancedb/native.d.ts \
+              --js ../lancedb/native.js \
+              --strip \
+              dist/
+      - name: Build
+        run: |
+          ${{ matrix.settings.pre_build }}
+          npx napi build --platform  --release --no-const-enum \
+              --features ${{ matrix.settings.features }} \
+              --target ${{ matrix.settings.target }} \
+              --dts ../lancedb/native.d.ts \
+              --js ../lancedb/native.js \
+              --strip \
+              $EXTRA_ARGS \
+              dist/
+        if: ${{ !matrix.settings.docker }}
+        shell: bash
+      - name: Upload artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: lancedb-${{ matrix.settings.target }}
+          path: nodejs/dist/*.node
+          if-no-files-found: error
+      # The generic files are the same in all distros so we just pick
+      # one to do the upload.
+      - name: Make generic artifacts
+        if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
+        run: npm run tsc
+      - name: Upload Generic Artifacts
+        if: ${{ matrix.settings.target == 'aarch64-apple-darwin' }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: nodejs-dist
+          path: |
+            nodejs/dist/*
+            !nodejs/dist/*.node
+  test-lancedb:
+    name: "Test: ${{ matrix.settings.target }} - node@${{ matrix.node }}"
+    needs:
+      - build-lancedb
+    strategy:
+      fail-fast: false
+      matrix:
+        settings:
+          # TODO: Get tests passing on Windows (failing from test tmpdir issue)
+          # - host: windows-latest
+          #   target: x86_64-pc-windows-msvc
+          - host: macos-latest
+            target: aarch64-apple-darwin
+          - target: x86_64-unknown-linux-gnu
+            host: ubuntu-latest
+          - target: aarch64-unknown-linux-gnu
+            host: buildjet-16vcpu-ubuntu-2204-arm
+        node:
+          - '20'
+    runs-on: ${{ matrix.settings.host }}
+    defaults:
+      run:
+        shell: bash
+        working-directory: nodejs
+    steps:
+      - uses: actions/checkout@v4
+      - name: Setup node
+        uses: actions/setup-node@v4
+        with:
+          node-version: ${{ matrix.node }}
+          cache: npm
+          cache-dependency-path: nodejs/package-lock.json
+      - name: Install dependencies
+        run: npm ci
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: lancedb-${{ matrix.settings.target }}
+          path: nodejs/dist/
+          # For testing purposes:
+          # run-id: 13982782871
+          # github-token: ${{ secrets.GITHUB_TOKEN }} # token with actions:read permissions on target repo
+      - uses: actions/download-artifact@v4
+        with:
+          name: nodejs-dist
+          path: nodejs/dist
+          # For testing purposes:
+          # github-token: ${{ secrets.GITHUB_TOKEN }} # token with actions:read permissions on target repo
+          # run-id: 13982782871
+      - name: List packages
+        run: ls -R dist
+      - name: Move built files
+        run: cp dist/native.d.ts dist/native.js dist/*.node lancedb/
+      - name: Test bindings
+        run: npm test
+  publish:
+    name: Publish
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        shell: bash
+        working-directory: nodejs
+    needs:
+      - test-lancedb
+    steps:
+      - uses: actions/checkout@v4
+      - name: Setup node
+        uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: npm
+          cache-dependency-path: nodejs/package-lock.json
+          registry-url: "https://registry.npmjs.org"
+      - name: Install dependencies
+        run: npm ci
+      - uses: actions/download-artifact@v4
+        with:
+          name: nodejs-dist
+          path: nodejs/dist
+          # For testing purposes:
+          # run-id: 13982782871
+          # github-token: ${{ secrets.GITHUB_TOKEN }} # token with actions:read permissions on target repo
+      - uses: actions/download-artifact@v4
+        name: Download arch-specific binaries
+        with:
+          pattern: lancedb-*
+          path: nodejs/nodejs-artifacts
+          merge-multiple: true
+          # For testing purposes:
+          # run-id: 13982782871
+          # github-token: ${{ secrets.GITHUB_TOKEN }} # token with actions:read permissions on target repo
+      - name: Display structure of downloaded files
+        run: find dist && find nodejs-artifacts
+      - name: Move artifacts
+        run: npx napi artifacts -d nodejs-artifacts
+      - name: List packages
+        run: find npm
+      - name: Publish
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.LANCEDB_NPM_REGISTRY_TOKEN }}
+          DRY_RUN: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+        run: |
+          ARGS="--access public"
+          if [[ $DRY_RUN == "true" ]]; then
+            ARGS="$ARGS --dry-run"
+          fi
+          if [[ $GITHUB_REF =~ refs/tags/v(.*)-beta.* ]]; then
+            ARGS="$ARGS --tag preview"
+          fi
+          npm publish $ARGS

.github/workflows/pypi-publish.yml

@@ -4,6 +4,11 @@ on:
   push:
     tags:
       - 'python-v*'
+  pull_request:
+    # This should trigger a dry run (we skip the final publish step)
+    paths:
+      - .github/workflows/pypi-publish.yml
+      - Cargo.toml # Change in dependency frequently breaks builds
 
 jobs:
   linux:
@@ -15,15 +20,21 @@ jobs:
           - platform: x86_64
             manylinux: "2_17"
             extra_args: ""
+            runner: ubuntu-22.04
           - platform: x86_64
             manylinux: "2_28"
             extra_args: "--features fp16kernels"
+            runner: ubuntu-22.04
           - platform: aarch64
-            manylinux: "2_24"
+            manylinux: "2_17"
             extra_args: ""
-          # We don't build fp16 kernels for aarch64, because it uses 
-          # cross compilation image, which doesn't have a new enough compiler.
-    runs-on: "ubuntu-22.04"
+            # For successful fat LTO builds, we need a large runner to avoid OOM errors.
+            runner: ubuntu-2404-8x-arm64
+          - platform: aarch64
+            manylinux: "2_28"
+            extra_args: "--features fp16kernels"
+            runner: ubuntu-2404-8x-arm64
+    runs-on: ${{ matrix.config.runner }}
     steps:
       - uses: actions/checkout@v4
         with:
@@ -40,6 +51,7 @@ jobs:
           arm-build: ${{ matrix.config.platform == 'aarch64' }}
           manylinux: ${{ matrix.config.manylinux }}
       - uses: ./.github/workflows/upload_wheel
+        if: startsWith(github.ref, 'refs/tags/python-v')
         with:
           pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
@@ -69,6 +81,7 @@ jobs:
           python-minor-version: 8
           args: "--release --strip --target ${{ matrix.config.target }} --features fp16kernels"
       - uses: ./.github/workflows/upload_wheel
+        if: startsWith(github.ref, 'refs/tags/python-v')
         with:
           pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
@@ -83,17 +96,19 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: 3.8
+          python-version: 3.12
       - uses: ./.github/workflows/build_windows_wheel
         with:
           python-minor-version: 8
           args: "--release --strip"
           vcpkg_token: ${{ secrets.VCPKG_GITHUB_PACKAGES }}
       - uses: ./.github/workflows/upload_wheel
+        if: startsWith(github.ref, 'refs/tags/python-v')
         with:
           pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
           fury_token: ${{ secrets.FURY_TOKEN }}
   gh-release:
+    if: startsWith(github.ref, 'refs/tags/python-v')
     runs-on: ubuntu-latest
     permissions:
       contents: write

.github/workflows/python.yml

@@ -13,6 +13,11 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
 
+env:
+  # Color output for pytest is off by default.
+  PYTEST_ADDOPTS: "--color=yes"
+  FORCE_COLOR: "1"
+
 jobs:
   lint:
     name: "Lint"
@@ -30,16 +35,17 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: "3.11"
+          python-version: "3.12"
       - name: Install ruff
         run: |
-          pip install ruff==0.2.2
+          pip install ruff==0.9.9
       - name: Format check
         run: ruff format --check .
       - name: Lint
-        run: ruff .
-  doctest:
-    name: "Doctest"
+        run: ruff check .
+
+  type-check:
+    name: "Type Check"
     timeout-minutes: 30
     runs-on: "ubuntu-22.04"
     defaults:
@@ -54,7 +60,36 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: "3.11"
+          python-version: "3.12"
+      - name: Install protobuf compiler
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler
+          pip install toml
+      - name: Install dependencies
+        run: |
+          python ../ci/parse_requirements.py pyproject.toml --extras dev,tests,embeddings > requirements.txt
+          pip install -r requirements.txt
+      - name: Run pyright
+        run: pyright
+
+  doctest:
+    name: "Doctest"
+    timeout-minutes: 30
+    runs-on: "ubuntu-24.04"
+    defaults:
+      run:
+        shell: bash
+        working-directory: python
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
           cache: "pip"
       - name: Install protobuf
         run: |
@@ -75,8 +110,8 @@ jobs:
     timeout-minutes: 30
     strategy:
       matrix:
-        python-minor-version: ["9", "11"]
-    runs-on: "ubuntu-22.04"
+        python-minor-version: ["9", "12"]
+    runs-on: "ubuntu-24.04"
     defaults:
       run:
         shell: bash
@@ -101,6 +136,10 @@ jobs:
       - uses: ./.github/workflows/run_tests
         with:
           integration: true
+      - name: Test without pylance or pandas
+        run: |
+          pip uninstall -y pylance pandas
+          pytest -vv python/tests/test_table.py
       # Make sure wheels are not included in the Rust cache
       - name: Delete wheels
         run: rm -rf target/wheels
@@ -127,7 +166,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: "3.11"
+          python-version: "3.12"
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: python
@@ -138,7 +177,7 @@ jobs:
         run: rm -rf target/wheels
   windows:
     name: "Windows: ${{ matrix.config.name }}"
-    timeout-minutes: 30
+    timeout-minutes: 60
     strategy:
       matrix:
         config:
@@ -157,7 +196,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: "3.11"
+          python-version: "3.12"
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: python
@@ -168,7 +207,7 @@ jobs:
         run: rm -rf target/wheels
   pydantic1x:
     timeout-minutes: 30
-    runs-on: "ubuntu-22.04"
+    runs-on: "ubuntu-24.04"
     defaults:
       run:
         shell: bash
@@ -189,6 +228,7 @@ jobs:
       - name: Install lancedb
         run: |
           pip install "pydantic<2"
+          pip install pyarrow==16
           pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests]
           pip install tantivy
       - name: Run tests

.github/workflows/run_tests/action.yml

@@ -24,8 +24,8 @@ runs:
     - name: pytest (with integration)
       shell: bash
       if: ${{ inputs.integration == 'true' }}
-      run: pytest -m "not slow" -x -v --durations=30 python/python/tests
+      run: pytest -m "not slow" -vv --durations=30 python/python/tests
     - name: pytest (no integration tests)
       shell: bash
       if: ${{ inputs.integration != 'true' }}
-      run: pytest -m "not slow and not s3_test" -x -v --durations=30 python/python/tests
+      run: pytest -m "not slow and not s3_test" -vv --durations=30 python/python/tests

.github/workflows/rust.yml

@@ -22,72 +22,116 @@ env:
   # "1" means line tables only, which is useful for panic tracebacks.
   RUSTFLAGS: "-C debuginfo=1"
   RUST_BACKTRACE: "1"
+  CARGO_INCREMENTAL: 0
 
 jobs:
   lint:
     timeout-minutes: 30
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-24.04
     defaults:
       run:
         shell: bash
-        working-directory: rust
     env:
       # Need up-to-date compilers for kernels
-      CC: gcc-12
-      CXX: g++-12
+      CC: clang-18
+      CXX: clang++-18
     steps:
-    - uses: actions/checkout@v4
-      with:
+      - uses: actions/checkout@v4
+        with:
           fetch-depth: 0
           lfs: true
-    - uses: Swatinem/rust-cache@v2
-      with:
-        workspaces: rust
-    - name: Install dependencies
-      run: |
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          components: rustfmt, clippy
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: rust
+      - name: Install dependencies
+        run: |
           sudo apt update
           sudo apt install -y protobuf-compiler libssl-dev
-    - name: Run format
-      run: cargo fmt --all -- --check
-    - name: Run clippy
-      run: cargo clippy --all --all-features -- -D warnings
+      - name: Run format
+        run: cargo fmt --all -- --check
+      - name: Run clippy
+        run: cargo clippy --workspace --tests --all-features -- -D warnings
+
+  build-no-lock:
+    runs-on: ubuntu-24.04
+    timeout-minutes: 30
+    env:
+      # Need up-to-date compilers for kernels
+      CC: clang
+      CXX: clang++
+    steps:
+      - uses: actions/checkout@v4
+      # Building without a lock file often requires the latest Rust version since downstream
+      # dependencies may have updated their minimum Rust version.
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          toolchain: "stable"
+      # Remove cargo.lock to force a fresh build
+      - name: Remove Cargo.lock
+        run: rm -f Cargo.lock
+      - uses: rui314/setup-mold@v1
+      - uses: Swatinem/rust-cache@v2
+      - name: Install dependencies
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler libssl-dev
+      - name: Build all
+        run: |
+          cargo build --benches --all-features --tests
+
   linux:
     timeout-minutes: 30
-    runs-on: ubuntu-22.04
+    # To build all features, we need more disk space than is available
+    # on the free OSS github runner. This is mostly due to the the
+    # sentence-transformers feature.
+    runs-on: ubuntu-2404-4x-x64
     defaults:
       run:
         shell: bash
         working-directory: rust
     env:
       # Need up-to-date compilers for kernels
-      CC: gcc-12
-      CXX: g++-12
+      CC: clang-18
+      CXX: clang++-18
     steps:
-    - uses: actions/checkout@v4
-      with:
+      - uses: actions/checkout@v4
+        with:
           fetch-depth: 0
           lfs: true
-    - uses: Swatinem/rust-cache@v2
-      with:
+      - uses: Swatinem/rust-cache@v2
+        with:
           workspaces: rust
-    - name: Install dependencies
-      run: |
-          sudo apt update
+      - name: Install dependencies
+        run: |
+          # This shaves 2 minutes off this step in CI. This doesn't seem to be
+          # necessary in standard runners, but it is in the 4x runners.
+          sudo rm /var/lib/man-db/auto-update
           sudo apt install -y protobuf-compiler libssl-dev
-    - name: Start S3 integration test environment
-      working-directory: .
-      run: docker compose up --detach --wait
-    - name: Build
-      run: cargo build --all-features
-    - name: Run tests
-      run: cargo test --all-features
-    - name: Run examples
-      run: cargo run --example simple
+      - uses: rui314/setup-mold@v1
+      - name: Make Swap
+        run: |
+          sudo fallocate -l 16G /swapfile
+          sudo chmod 600 /swapfile
+          sudo mkswap /swapfile
+          sudo swapon /swapfile
+      - name: Start S3 integration test environment
+        working-directory: .
+        run: docker compose up --detach --wait
+      - name: Build
+        run: cargo build --all-features --tests --locked --examples
+      - name: Run tests
+        run: cargo test --all-features --locked
+      - name: Run examples
+        run: cargo run --example simple --locked
+
   macos:
     timeout-minutes: 30
     strategy:
       matrix:
-        mac-runner: [ "macos-13", "macos-14" ]
+        mac-runner: ["macos-13", "macos-14"]
     runs-on: "${{ matrix.mac-runner }}"
     defaults:
       run:
@@ -96,8 +140,8 @@ jobs:
     steps:
       - uses: actions/checkout@v4
         with:
-            fetch-depth: 0
-            lfs: true
+          fetch-depth: 0
+          lfs: true
       - name: CPU features
         run: sysctl -a | grep cpu
       - uses: Swatinem/rust-cache@v2
@@ -105,30 +149,78 @@ jobs:
           workspaces: rust
       - name: Install dependencies
         run: brew install protobuf
-      - name: Build
-        run: cargo build --all-features
       - name: Run tests
-        # Run with everything except the integration tests.
-        run: cargo test --features remote,fp16kernels
+        run: |
+          # Don't run the s3 integration tests since docker isn't available
+          # on this image.
+          ALL_FEATURES=`cargo metadata --format-version=1 --no-deps \
+            | jq -r '.packages[] | .features | keys | .[]' \
+            | grep -v s3-test | sort | uniq | paste -s -d "," -`
+          cargo test --features $ALL_FEATURES --locked
+
   windows:
     runs-on: windows-2022
+    strategy:
+      matrix:
+        target:
+          - x86_64-pc-windows-msvc
+          - aarch64-pc-windows-msvc
+    defaults:
+      run:
+        working-directory: rust/lancedb
     steps:
       - uses: actions/checkout@v4
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: rust
       - name: Install Protoc v21.12
-        working-directory: C:\
+        run: choco install --no-progress protoc
+      - name: Build
         run: |
-          New-Item -Path 'C:\protoc' -ItemType Directory
-          Set-Location C:\protoc
-          Invoke-WebRequest https://github.com/protocolbuffers/protobuf/releases/download/v21.12/protoc-21.12-win64.zip -OutFile C:\protoc\protoc.zip
-          7z x protoc.zip
-          Add-Content $env:GITHUB_PATH "C:\protoc\bin"
-        shell: powershell
+          rustup target add ${{ matrix.target }}
+          $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
+          cargo build --features remote --tests --locked --target ${{ matrix.target }}
       - name: Run tests
+        # Can only run tests when target matches host
+        if: ${{ matrix.target == 'x86_64-pc-windows-msvc' }}
         run: |
           $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
-          cargo build
-          cargo test
-  
+          cargo test --features remote --locked
+
+  msrv:
+    # Check the minimum supported Rust version
+    name: MSRV Check - Rust v${{ matrix.msrv }}
+    runs-on: ubuntu-24.04
+    strategy:
+      matrix:
+        msrv: ["1.78.0"] # This should match up with rust-version in Cargo.toml
+    env:
+      # Need up-to-date compilers for kernels
+      CC: clang-18
+      CXX: clang++-18
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Install dependencies
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler libssl-dev
+      - name: Install ${{ matrix.msrv }}
+        uses: dtolnay/rust-toolchain@master
+        with:
+          toolchain: ${{ matrix.msrv }}
+      - name: Downgrade  dependencies
+        # These packages have newer requirements for MSRV
+        run: |
+          cargo update -p aws-sdk-bedrockruntime --precise 1.64.0
+          cargo update -p aws-sdk-dynamodb --precise 1.55.0
+          cargo update -p aws-config --precise 1.5.10
+          cargo update -p aws-sdk-kms --precise 1.51.0
+          cargo update -p aws-sdk-s3 --precise 1.65.0
+          cargo update -p aws-sdk-sso --precise 1.50.0
+          cargo update -p aws-sdk-ssooidc --precise 1.51.0
+          cargo update -p aws-sdk-sts --precise 1.51.0
+          cargo update -p home --precise 0.5.9
+      - name: cargo +${{ matrix.msrv }} check
+        run: cargo check --workspace --tests --benches --all-features

.github/workflows/update_package_lock/action.yml

@@ -1,33 +0,0 @@
-name: update_package_lock
-description: "Update node's package.lock"
-
-inputs:
-  github_token:
-    required: true
-    description: "github token for the repo"
-
-runs:
-  using: "composite"
-  steps:
-    - uses: actions/setup-node@v3
-      with:
-        node-version: 20
-    - name: Set git configs
-      shell: bash
-      run: |
-        git config user.name 'Lance Release'
-        git config user.email 'lance-dev@lancedb.com'
-    - name: Update package-lock.json file
-      working-directory: ./node
-      run: |
-        npm install
-        git add package-lock.json
-        git commit -m "Updating package-lock.json"
-      shell: bash
-    - name: Push changes
-      if: ${{ inputs.dry_run }} == "false"
-      uses: ad-m/github-push-action@master
-      with:
-        github_token: ${{ inputs.github_token }}
-        branch: main
-        tags: true

.github/workflows/update_package_lock_nodejs/action.yml

@@ -1,33 +0,0 @@
-name: update_package_lock_nodejs
-description: "Update nodejs's package.lock"
-
-inputs:
-  github_token:
-    required: true
-    description: "github token for the repo"
-
-runs:
-  using: "composite"
-  steps:
-    - uses: actions/setup-node@v3
-      with:
-        node-version: 20
-    - name: Set git configs
-      shell: bash
-      run: |
-        git config user.name 'Lance Release'
-        git config user.email 'lance-dev@lancedb.com'
-    - name: Update package-lock.json file
-      working-directory: ./nodejs
-      run: |
-        npm install
-        git add package-lock.json
-        git commit -m "Updating package-lock.json"
-      shell: bash
-    - name: Push changes
-      if: ${{ inputs.dry_run }} == "false"
-      uses: ad-m/github-push-action@master
-      with:
-        github_token: ${{ inputs.github_token }}
-        branch: main
-        tags: true

.github/workflows/upload_wheel/action.yml

@@ -17,11 +17,12 @@ runs:
     run: |
       python -m pip install --upgrade pip
       pip install twine
+      python3 -m pip install --upgrade pkginfo
   - name: Choose repo
     shell: bash
     id: choose_repo
     run: |
-      if [ ${{ github.ref }} == "*beta*" ]; then
+      if [[ ${{ github.ref }} == *beta* ]]; then
         echo "repo=fury" >> $GITHUB_OUTPUT
       else
         echo "repo=pypi" >> $GITHUB_OUTPUT
@@ -32,7 +33,7 @@ runs:
       FURY_TOKEN: ${{ inputs.fury_token }}
       PYPI_TOKEN: ${{ inputs.pypi_token }}
     run: |
-      if [ ${{ steps.choose_repo.outputs.repo }} == "fury" ]; then
+      if [[ ${{ steps.choose_repo.outputs.repo }} == fury ]]; then
         WHEEL=$(ls target/wheels/lancedb-*.whl 2> /dev/null | head -n 1)
         echo "Uploading $WHEEL to Fury"
         curl -f -F package=@$WHEEL https://$FURY_TOKEN@push.fury.io/lancedb/

.gitignore

@@ -9,7 +9,6 @@ venv
 .vscode
 .zed
 rust/target
-rust/Cargo.lock
 
 site
 
@@ -32,9 +31,6 @@ python/dist
 *.node
 **/node_modules
 **/.DS_Store
-node/dist
-node/examples/**/package-lock.json
-node/examples/**/dist
 nodejs/lancedb/native*
 dist
 
@@ -42,5 +38,3 @@ dist
 target
 
 **/sccache.log
-
-Cargo.lock

.pre-commit-config.yaml

@@ -1,21 +1,27 @@
 repos:
--   repo: https://github.com/pre-commit/pre-commit-hooks
+  - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v3.2.0
     hooks:
-    -   id: check-yaml
-    -   id: end-of-file-fixer
-    -   id: trailing-whitespace
--   repo: https://github.com/astral-sh/ruff-pre-commit
+      - id: check-yaml
+      - id: end-of-file-fixer
+      - id: trailing-whitespace
+  - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.2.2
+    rev: v0.9.9
     hooks:
-    - id: ruff
-- repo: local
-  hooks:
-    - id: local-biome-check
-      name: biome check
-      entry: npx @biomejs/biome@1.8.3 check --config-path nodejs/biome.json nodejs/
-      language: system
-      types: [text]
-      files: "nodejs/.*"
-      exclude: nodejs/lancedb/native.d.ts|nodejs/dist/.*|nodejs/examples/.*
+      - id: ruff
+  # - repo: https://github.com/RobertCraigie/pyright-python
+  #   rev: v1.1.395
+  #   hooks:
+  #     - id: pyright
+  #       args: ["--project", "python"]
+  #       additional_dependencies: [pyarrow-stubs]
+  - repo: local
+    hooks:
+      - id: local-biome-check
+        name: biome check
+        entry: npx @biomejs/biome@1.8.3 check --config-path nodejs/biome.json nodejs/
+        language: system
+        types: [text]
+        files: "nodejs/.*"
+        exclude: nodejs/lancedb/native.d.ts|nodejs/dist/.*|nodejs/examples/.*

CLAUDE.md

@@ -0,0 +1,22 @@
+LanceDB is a database designed for retrieval, including vector, full-text, and hybrid search.
+It is a wrapper around Lance. There are two backends: local (in-process like SQLite) and
+remote (against LanceDB Cloud).
+
+The core of LanceDB is written in Rust. There are bindings in Python, Typescript, and Java.
+
+Project layout:
+
+* `rust/lancedb`: The LanceDB core Rust implementation.
+* `python`: The Python bindings, using PyO3.
+* `nodejs`: The Typescript bindings, using napi-rs
+* `java`: The Java bindings
+
+Common commands:
+
+* Check for compiler errors: `cargo check --features remote --tests --examples`
+* Run tests: `cargo test --features remote --tests`
+* Run specific test: `cargo test --features remote -p <package_name> --test <test_name>`
+* Lint: `cargo clippy --features remote --tests --examples`
+* Format: `cargo fmt --all`
+
+Before committing changes, run formatting.

CONTRIBUTING.md

@@ -0,0 +1,78 @@
+# Contributing to LanceDB
+
+LanceDB is an open-source project and we welcome contributions from the community.
+This document outlines the process for contributing to LanceDB.
+
+## Reporting Issues
+
+If you encounter a bug or have a feature request, please open an issue on the
+[GitHub issue tracker](https://github.com/lancedb/lancedb).
+
+## Picking an issue
+
+We track issues on the GitHub issue tracker. If you are looking for something to
+work on, check the [good first issue](https://github.com/lancedb/lancedb/contribute) label. These issues are typically the best described and have the smallest scope.
+
+If there's an issue you are interested in working on, please leave a comment on the issue. This will help us avoid duplicate work. Additionally, if you have questions about the issue, please ask them in the issue comments. We are happy to provide guidance on how to approach the issue.
+
+## Configuring Git
+
+First, fork the repository on GitHub, then clone your fork:
+
+```bash
+git clone https://github.com/<username>/lancedb.git
+cd lancedb
+```
+
+Then add the main repository as a remote:
+
+```bash
+git remote add upstream https://github.com/lancedb/lancedb.git
+git fetch upstream
+```
+
+## Setting up your development environment
+
+We have development environments for Python, Typescript, and Java. Each environment has its own setup instructions.
+
+* [Python](python/CONTRIBUTING.md)
+* [Typescript](nodejs/CONTRIBUTING.md)
+<!-- TODO: add Java contributing guide -->
+* [Documentation](docs/README.md)
+
+
+## Best practices for pull requests
+
+For the best chance of having your pull request accepted, please follow these guidelines:
+
+1. Unit test all bug fixes and new features. Your code will not be merged if it
+   doesn't have tests.
+1. If you change the public API, update the documentation in the `docs` directory.
+1. Aim to minimize the number of changes in each pull request. Keep to solving
+   one problem at a time, when possible.
+1. Before marking a pull request ready-for-review, do a self review of your code.
+   Is it clear why you are making the changes? Are the changes easy to understand?
+1. Use [conventional commit messages](https://www.conventionalcommits.org/en/) as pull request titles. Examples:
+    * New feature: `feat: adding foo API`
+    * Bug fix: `fix: issue with foo API`
+    * Documentation change: `docs: adding foo API documentation`
+1. If your pull request is a work in progress, leave the pull request as a draft.
+   We will assume the pull request is ready for review when it is opened.
+1. When writing tests, test the error cases. Make sure they have understandable
+   error messages.
+
+## Project structure
+
+The core library is written in Rust. The Python, Typescript, and Java libraries
+are wrappers around the Rust library.
+
+* `src/lancedb`: Rust library source code
+* `python`: Python package source code
+* `nodejs`: Typescript package source code
+* `node`: **Deprecated** Typescript package source code
+* `java`: Java package source code
+* `docs`: Documentation source code
+
+## Release process
+
+For information on the release process, see: [release_process.md](release_process.md)

Cargo.toml

@@ -1,11 +1,5 @@
 [workspace]
-members = [
-    "rust/ffi/node",
-    "rust/lancedb",
-    "nodejs",
-    "python",
-    "java/core/lancedb-jni",
-]
+members = ["rust/lancedb", "nodejs", "python", "java/core/lancedb-jni"]
 # Python package needs to be built by maturin.
 exclude = ["python"]
 resolver = "2"
@@ -18,34 +12,54 @@ repository = "https://github.com/lancedb/lancedb"
 description = "Serverless, low-latency vector database for AI applications"
 keywords = ["lancedb", "lance", "database", "vector", "search"]
 categories = ["database-implementations"]
+rust-version = "1.78.0"
 
 [workspace.dependencies]
-lance = { "version" = "=0.14.1", "features" = ["dynamodb"] }
-lance-index = { "version" = "=0.14.1" }
-lance-linalg = { "version" = "=0.14.1" }
-lance-testing = { "version" = "=0.14.1" }
-lance-datafusion = { "version" = "=0.14.1" }
+lance = { "version" = "=0.33.0", "features" = ["dynamodb"] }
+lance-io = "=0.33.0"
+lance-index = "=0.33.0"
+lance-linalg = "=0.33.0"
+lance-table = "=0.33.0"
+lance-testing = "=0.33.0"
+lance-datafusion = "=0.33.0"
+lance-encoding = "=0.33.0"
 # Note that this one does not include pyarrow
-arrow = { version = "51.0", optional = false }
-arrow-array = "51.0"
-arrow-data = "51.0"
-arrow-ipc = "51.0"
-arrow-ord = "51.0"
-arrow-schema = "51.0"
-arrow-arith = "51.0"
-arrow-cast = "51.0"
+arrow = { version = "55.1", optional = false }
+arrow-array = "55.1"
+arrow-data = "55.1"
+arrow-ipc = "55.1"
+arrow-ord = "55.1"
+arrow-schema = "55.1"
+arrow-arith = "55.1"
+arrow-cast = "55.1"
 async-trait = "0"
-chrono = "0.4.35"
-datafusion-physical-plan = "37.1"
-half = { "version" = "=2.4.1", default-features = false, features = [
+datafusion = { version = "48.0", default-features = false }
+datafusion-catalog = "48.0"
+datafusion-common = { version = "48.0", default-features = false }
+datafusion-execution = "48.0"
+datafusion-expr = "48.0"
+datafusion-physical-plan = "48.0"
+env_logger = "0.11"
+half = { "version" = "2.6.0", default-features = false, features = [
     "num-traits",
 ] }
 futures = "0"
 log = "0.4"
-object_store = "0.9.0"
+moka = { version = "0.12", features = ["future"] }
+object_store = "0.12.0"
 pin-project = "1.0.7"
-snafu = "0.7.4"
+snafu = "0.8"
 url = "2"
 num-traits = "0.2"
+rand = "0.9"
 regex = "1.10"
 lazy_static = "1"
+semver = "1.0.25"
+crunchy = "0.2.4"
+# Temporary pins to work around downstream issues
+# https://github.com/apache/arrow-rs/commit/2fddf85afcd20110ce783ed5b4cdeb82293da30b
+chrono = "=0.4.41"
+# https://github.com/RustCrypto/formats/issues/1684
+base64ct = "=1.6.0"
+# Workaround for: https://github.com/Lokathor/bytemuck/issues/306
+bytemuck_derive = ">=1.8.1, <1.9.0"

README.md

@@ -1,87 +1,97 @@
+<a href="https://cloud.lancedb.com" target="_blank">
+  <img src="https://github.com/user-attachments/assets/92dad0a2-2a37-4ce1-b783-0d1b4f30a00c" alt="LanceDB Cloud Public Beta" width="100%" style="max-width: 100%;">
+</a>
 <div align="center">
-<p align="center">
 
-<img width="275" alt="LanceDB Logo" src="https://github.com/lancedb/lancedb/assets/5846846/37d7c7ad-c2fd-4f56-9f16-fffb0d17c73a">
+[![LanceDB](docs/src/assets/hero-header.png)](https://lancedb.com)
+[![Website](https://img.shields.io/badge/-Website-100000?style=for-the-badge&labelColor=645cfb&color=645cfb)](https://lancedb.com/)
+[![Blog](https://img.shields.io/badge/Blog-100000?style=for-the-badge&labelColor=645cfb&color=645cfb)](https://blog.lancedb.com/)
+[![Discord](https://img.shields.io/badge/-Discord-100000?style=for-the-badge&logo=discord&logoColor=white&labelColor=645cfb&color=645cfb)](https://discord.gg/zMM32dvNtd)
+[![Twitter](https://img.shields.io/badge/-Twitter-100000?style=for-the-badge&logo=x&logoColor=white&labelColor=645cfb&color=645cfb)](https://twitter.com/lancedb)
+[![LinkedIn](https://img.shields.io/badge/-LinkedIn-100000?style=for-the-badge&logo=linkedin&logoColor=white&labelColor=645cfb&color=645cfb)](https://www.linkedin.com/company/lancedb/)
 
-**Developer-friendly, database for multimodal AI**
 
-<a href='https://github.com/lancedb/vectordb-recipes/tree/main' target="_blank"><img alt='LanceDB' src='https://img.shields.io/badge/VectorDB_Recipes-100000?style=for-the-badge&logo=LanceDB&logoColor=white&labelColor=645cfb&color=645cfb'/></a>
-<a href='https://lancedb.github.io/lancedb/' target="_blank"><img alt='lancdb' src='https://img.shields.io/badge/DOCS-100000?style=for-the-badge&logo=lancdb&logoColor=white&labelColor=645cfb&color=645cfb'/></a>
-[![Blog](https://img.shields.io/badge/Blog-12100E?style=for-the-badge&logoColor=white)](https://blog.lancedb.com/) 
-[![Discord](https://img.shields.io/badge/Discord-%235865F2.svg?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/zMM32dvNtd) 
-[![Twitter](https://img.shields.io/badge/Twitter-%231DA1F2.svg?style=for-the-badge&logo=Twitter&logoColor=white)](https://twitter.com/lancedb)
+<img src="docs/src/assets/lancedb.png" alt="LanceDB" width="50%">
 
-</p>
+# **The Multimodal AI Lakehouse**
 
-<img max-width="750px" alt="LanceDB Multimodal Search" src="https://github.com/lancedb/lancedb/assets/917119/09c5afc5-7816-4687-bae4-f2ca194426ec">
+[**How to Install** ](#how-to-install) ✦ [**Detailed Documentation**](https://lancedb.github.io/lancedb/) ✦ [**Tutorials and Recipes**](https://github.com/lancedb/vectordb-recipes/tree/main) ✦  [**Contributors**](#contributors) 
+
+**The ultimate multimodal data platform for AI/ML applications.** 
+
+LanceDB is designed for fast, scalable, and production-ready vector search. It is built on top of the Lance columnar format. You can store, index, and search over petabytes of multimodal data and vectors with ease. 
+LanceDB is a central location where developers can build, train and analyze their AI workloads.
 
-</p>
 </div>
 
-<hr />
+<br>
 
-LanceDB is an open-source database for vector-search built with persistent storage, which greatly simplifies retrieval, filtering and management of embeddings.
+## **Demo: Multimodal Search by Keyword, Vector or with SQL**
+<img max-width="750px" alt="LanceDB Multimodal Search" src="https://github.com/lancedb/lancedb/assets/917119/09c5afc5-7816-4687-bae4-f2ca194426ec">
 
-The key features of LanceDB include:
+## **Star LanceDB to get updates!**
 
-* Production-scale vector search with no servers to manage.
+<details>
+<summary>⭐ Click here ⭐  to see how fast we're growing!</summary>
+<picture>
+  <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=lancedb/lancedb&theme=dark&type=Date">
+  <img width="100%" src="https://api.star-history.com/svg?repos=lancedb/lancedb&theme=dark&type=Date">
+</picture>
+</details>
 
-* Store, query and filter vectors, metadata and multi-modal data (text, images, videos, point clouds, and more).
+## **Key Features**:
 
-* Support for vector similarity search, full-text search and SQL.
+- **Fast Vector Search**: Search billions of vectors in milliseconds with state-of-the-art indexing.
+- **Comprehensive Search**: Support for vector similarity search, full-text search and SQL.
+- **Multimodal Support**: Store, query and filter vectors, metadata and multimodal data (text, images, videos, point clouds, and more).
+- **Advanced Features**: Zero-copy, automatic versioning, manage versions of your data without needing extra infrastructure. GPU support in building vector index.
 
-* Native Python and Javascript/Typescript support.
+### **Products**:
+- **Open Source & Local**: 100% open source, runs locally or in your cloud. No vendor lock-in.
+- **Cloud and Enterprise**: Production-scale vector search with no servers to manage. Complete data sovereignty and security.
 
-* Zero-copy, automatic versioning, manage versions of your data without needing extra infrastructure.
+### **Ecosystem**:
+- **Columnar Storage**: Built on the Lance columnar format for efficient storage and analytics.
+- **Seamless Integration**: Python, Node.js, Rust, and REST APIs for easy integration. Native Python and Javascript/Typescript support.
+- **Rich Ecosystem**: Integrations with [**LangChain** 🦜️🔗](https://python.langchain.com/docs/integrations/vectorstores/lancedb/), [**LlamaIndex** 🦙](https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html), Apache-Arrow, Pandas, Polars, DuckDB and more on the way.
 
-* GPU support in building vector index(*).
+## **How to Install**:
 
-* Ecosystem integrations with [LangChain 🦜️🔗](https://python.langchain.com/docs/integrations/vectorstores/lancedb/), [LlamaIndex 🦙](https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html), Apache-Arrow, Pandas, Polars, DuckDB and more on the way.
+Follow the [Quickstart](https://lancedb.github.io/lancedb/basic/) doc to set up LanceDB locally. 
 
-LanceDB's core is written in Rust 🦀 and is built using <a href="https://github.com/lancedb/lance">Lance</a>, an open-source columnar format designed for performant ML workloads.
+**API & SDK:** We also support Python, Typescript and Rust SDKs
 
-## Quick Start
+| Interface | Documentation |
+|-----------|---------------|
+| Python SDK | https://lancedb.github.io/lancedb/python/python/ |
+| Typescript SDK | https://lancedb.github.io/lancedb/js/globals/ |
+| Rust SDK | https://docs.rs/lancedb/latest/lancedb/index.html |
+| REST API | https://docs.lancedb.com/api-reference/introduction |
 
-**Javascript**
-```shell
-npm install vectordb
-```
+## **Join Us and Contribute**
 
-```javascript
-const lancedb = require('vectordb');
-const db = await lancedb.connect('data/sample-lancedb');
+We welcome contributions from everyone! Whether you're a developer, researcher, or just someone who wants to help out. 
 
-const table = await db.createTable({
-  name: 'vectors',
-  data:  [
-    { id: 1, vector: [0.1, 0.2], item: "foo", price: 10 },
-    { id: 2, vector: [1.1, 1.2], item: "bar", price: 50 }
-  ]
-})
+If you have any suggestions or feature requests, please feel free to open an issue on GitHub or discuss it on our [**Discord**](https://discord.gg/G5DcmnZWKB) server.
 
-const query = table.search([0.1, 0.3]).limit(2);
-const results = await query.execute();
+[**Check out the GitHub Issues**](https://github.com/lancedb/lancedb/issues) if you would like to work on the features that are planned for the future. If you have any suggestions or feature requests, please feel free to open an issue on GitHub. 
 
-// You can also search for rows by specific criteria without involving a vector search.
-const rowsByCriteria = await table.search(undefined).where("price >= 10").execute();
-```
+## **Contributors**
 
-**Python**
-```shell
-pip install lancedb
-```
+<a href="https://github.com/lancedb/lancedb/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=lancedb/lancedb" />
+</a>
 
-```python
-import lancedb
 
-uri = "data/sample-lancedb"
-db = lancedb.connect(uri)
-table = db.create_table("my_table",
-                         data=[{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-                               {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
-result = table.search([100, 100]).limit(2).to_pandas()
-```
+## **Stay in Touch With Us**
+<div align="center">
 
-## Blogs, Tutorials & Videos
-* 📈 <a href="https://blog.lancedb.com/benchmarking-random-access-in-lance/">2000x better performance with Lance over Parquet</a>
-* 🤖 <a href="https://github.com/lancedb/lancedb/blob/main/docs/src/notebooks/youtube_transcript_search.ipynb">Build a question and answer bot with LanceDB</a>
+</br>
+
+[![Website](https://img.shields.io/badge/-Website-100000?style=for-the-badge&labelColor=645cfb&color=645cfb)](https://lancedb.com/)
+[![Blog](https://img.shields.io/badge/Blog-100000?style=for-the-badge&labelColor=645cfb&color=645cfb)](https://blog.lancedb.com/)
+[![Discord](https://img.shields.io/badge/-Discord-100000?style=for-the-badge&logo=discord&logoColor=white&labelColor=645cfb&color=645cfb)](https://discord.gg/zMM32dvNtd)
+[![Twitter](https://img.shields.io/badge/-Twitter-100000?style=for-the-badge&logo=x&logoColor=white&labelColor=645cfb&color=645cfb)](https://twitter.com/lancedb)
+[![LinkedIn](https://img.shields.io/badge/-LinkedIn-100000?style=for-the-badge&logo=linkedin&logoColor=white&labelColor=645cfb&color=645cfb)](https://www.linkedin.com/company/lancedb/)
+
+</div>

ci/build_linux_artifacts.sh

@@ -1,21 +0,0 @@
-#!/bin/bash
-set -e
-ARCH=${1:-x86_64}
-
-# We pass down the current user so that when we later mount the local files 
-# into the container, the files are accessible by the current user.
-pushd ci/manylinux_node
-docker build \
-    -t lancedb-node-manylinux \
-    --build-arg="ARCH=$ARCH" \
-    --build-arg="DOCKER_USER=$(id -u)" \
-    --progress=plain \
-    .
-popd
-
-# We turn on memory swap to avoid OOM killer
-docker run \
-    -v $(pwd):/io -w /io \
-    --memory-swap=-1 \
-    lancedb-node-manylinux \
-    bash ci/manylinux_node/build.sh $ARCH

ci/build_linux_artifacts_nodejs.sh

@@ -1,21 +0,0 @@
-#!/bin/bash
-set -e
-ARCH=${1:-x86_64}
-
-# We pass down the current user so that when we later mount the local files
-# into the container, the files are accessible by the current user.
-pushd ci/manylinux_nodejs
-docker build \
-    -t lancedb-nodejs-manylinux \
-    --build-arg="ARCH=$ARCH" \
-    --build-arg="DOCKER_USER=$(id -u)" \
-    --progress=plain \
-    .
-popd
-
-# We turn on memory swap to avoid OOM killer
-docker run \
-    -v $(pwd):/io -w /io \
-    --memory-swap=-1 \
-    lancedb-nodejs-manylinux \
-    bash ci/manylinux_nodejs/build.sh $ARCH

ci/build_macos_artifacts.sh

@@ -1,34 +0,0 @@
-# Builds the macOS artifacts (node binaries).
-# Usage: ./ci/build_macos_artifacts.sh [target]
-# Targets supported: x86_64-apple-darwin aarch64-apple-darwin
-set -e
-
-prebuild_rust() {
-    # Building here for the sake of easier debugging.
-    pushd rust/ffi/node
-    echo "Building rust library for $1"
-    export RUST_BACKTRACE=1
-    cargo build --release --target $1
-    popd
-}
-
-build_node_binaries() {
-    pushd node
-    echo "Building node library for $1"
-    npm run build-release -- --target $1
-    npm run pack-build -- --target $1
-    popd
-}
-
-if [ -n "$1" ]; then
-    targets=$1
-else
-    targets="x86_64-apple-darwin aarch64-apple-darwin"
-fi
-
-echo "Building artifacts for targets: $targets"
-for target in $targets
-    do
-    prebuild_rust $target
-    build_node_binaries $target
-done

ci/build_macos_artifacts_nodejs.sh

@@ -1,34 +0,0 @@
-# Builds the macOS artifacts (nodejs binaries).
-# Usage: ./ci/build_macos_artifacts_nodejs.sh [target]
-# Targets supported: x86_64-apple-darwin aarch64-apple-darwin
-set -e
-
-prebuild_rust() {
-    # Building here for the sake of easier debugging.
-    pushd rust/lancedb
-    echo "Building rust library for $1"
-    export RUST_BACKTRACE=1
-    cargo build --release --target $1
-    popd
-}
-
-build_node_binaries() {
-    pushd nodejs
-    echo "Building nodejs library for $1"
-    export RUST_TARGET=$1
-    npm run build-release
-    popd
-}
-
-if [ -n "$1" ]; then
-    targets=$1
-else
-    targets="x86_64-apple-darwin aarch64-apple-darwin"
-fi
-
-echo "Building artifacts for targets: $targets"
-for target in $targets
-    do
-    prebuild_rust $target
-    build_node_binaries $target
-done

ci/build_windows_artifacts.ps1

@@ -1,41 +0,0 @@
-# Builds the Windows artifacts (node binaries).
-# Usage:  .\ci\build_windows_artifacts.ps1 [target]
-# Targets supported:
-# - x86_64-pc-windows-msvc
-# - i686-pc-windows-msvc
-
-function Prebuild-Rust {
-    param (
-        [string]$target
-    )
-
-    # Building here for the sake of easier debugging.
-    Push-Location -Path "rust/ffi/node"
-    Write-Host "Building rust library for $target"
-    $env:RUST_BACKTRACE=1
-    cargo build --release --target $target
-    Pop-Location
-}
-
-function Build-NodeBinaries {
-    param (
-        [string]$target
-    )
-
-    Push-Location -Path "node"
-    Write-Host "Building node library for $target"
-    npm run build-release -- --target $target
-    npm run pack-build -- --target $target
-    Pop-Location
-}
-
-$targets = $args[0]
-if (-not $targets) {
-    $targets = "x86_64-pc-windows-msvc"
-}
-
-Write-Host "Building artifacts for targets: $targets"
-foreach ($target in $targets) {
-    Prebuild-Rust $target
-    Build-NodeBinaries $target
-}

ci/build_windows_artifacts_nodejs.ps1

@@ -1,41 +0,0 @@
-# Builds the Windows artifacts (nodejs binaries).
-# Usage:  .\ci\build_windows_artifacts_nodejs.ps1 [target]
-# Targets supported:
-# - x86_64-pc-windows-msvc
-# - i686-pc-windows-msvc
-
-function Prebuild-Rust {
-    param (
-        [string]$target
-    )
-
-    # Building here for the sake of easier debugging.
-    Push-Location -Path "rust/lancedb"
-    Write-Host "Building rust library for $target"
-    $env:RUST_BACKTRACE=1
-    cargo build --release --target $target
-    Pop-Location
-}
-
-function Build-NodeBinaries {
-    param (
-        [string]$target
-    )
-
-    Push-Location -Path "nodejs"
-    Write-Host "Building nodejs library for $target"
-    $env:RUST_TARGET=$target
-    npm run build-release
-    Pop-Location
-}
-
-$targets = $args[0]
-if (-not $targets) {
-    $targets = "x86_64-pc-windows-msvc"
-}
-
-Write-Host "Building artifacts for targets: $targets"
-foreach ($target in $targets) {
-    Prebuild-Rust $target
-    Build-NodeBinaries $target
-}

ci/manylinux_node/Dockerfile

@@ -1,31 +0,0 @@
-# Many linux dockerfile with Rust, Node, and Lance dependencies installed.
-# This container allows building the node modules native libraries in an 
-# environment with a very old glibc, so that we are compatible with a wide
-# range of linux distributions.
-ARG ARCH=x86_64
-
-FROM quay.io/pypa/manylinux2014_${ARCH}
-
-ARG ARCH=x86_64
-ARG DOCKER_USER=default_user
-
-# Install static openssl
-COPY install_openssl.sh install_openssl.sh
-RUN ./install_openssl.sh ${ARCH} > /dev/null
-
-# Protobuf is also installed as root.
-COPY install_protobuf.sh install_protobuf.sh
-RUN ./install_protobuf.sh ${ARCH}
-
-ENV DOCKER_USER=${DOCKER_USER}
-# Create a group and user
-RUN echo ${ARCH} && adduser --user-group --create-home --uid ${DOCKER_USER} build_user
-
-# We switch to the user to install Rust and Node, since those like to be 
-# installed at the user level.
-USER ${DOCKER_USER}
-
-COPY prepare_manylinux_node.sh prepare_manylinux_node.sh
-RUN cp /prepare_manylinux_node.sh $HOME/ && \
-    cd $HOME && \
-    ./prepare_manylinux_node.sh ${ARCH}

ci/manylinux_node/build.sh

@@ -1,19 +0,0 @@
-#!/bin/bash
-# Builds the node module for manylinux. Invoked by ci/build_linux_artifacts.sh.
-set -e
-ARCH=${1:-x86_64}
-
-if [ "$ARCH" = "x86_64" ]; then
-    export OPENSSL_LIB_DIR=/usr/local/lib64/
-else 
-    export OPENSSL_LIB_DIR=/usr/local/lib/
-fi
-export OPENSSL_STATIC=1
-export OPENSSL_INCLUDE_DIR=/usr/local/include/openssl
-
-source $HOME/.bashrc
-
-cd node
-npm ci
-npm run build-release
-npm run pack-build

ci/manylinux_node/install_openssl.sh

@@ -1,26 +0,0 @@
-#!/bin/bash
-# Builds openssl from source so we can statically link to it
-
-# this is to avoid the error we get with the system installation:
-# /usr/bin/ld: <library>: version node not found for symbol SSLeay@@OPENSSL_1.0.1
-# /usr/bin/ld: failed to set dynamic section sizes: Bad value
-set -e
-
-git clone -b OpenSSL_1_1_1u \
-    --single-branch \
-    https://github.com/openssl/openssl.git
-
-pushd openssl
-
-if [[ $1 == x86_64* ]]; then
-    ARCH=linux-x86_64
-else
-    # gnu target
-    ARCH=linux-aarch64
-fi
-
-./Configure no-shared $ARCH
-
-make
-
-make install

ci/manylinux_node/install_protobuf.sh

@@ -1,15 +0,0 @@
-#!/bin/bash
-# Installs protobuf compiler. Should be run as root.
-set -e
-
-if [[ $1 == x86_64* ]]; then
-    ARCH=x86_64
-else
-    # gnu target
-    ARCH=aarch_64
-fi
-
-PB_REL=https://github.com/protocolbuffers/protobuf/releases
-PB_VERSION=23.1
-curl -LO $PB_REL/download/v$PB_VERSION/protoc-$PB_VERSION-linux-$ARCH.zip
-unzip protoc-$PB_VERSION-linux-$ARCH.zip -d /usr/local

ci/manylinux_node/prepare_manylinux_node.sh

@@ -1,21 +0,0 @@
-#!/bin/bash
-set -e
-
-install_node() {
-    echo "Installing node..."
-
-    curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.34.0/install.sh | bash
-
-    source "$HOME"/.bashrc
-
-    nvm install --no-progress 16
-}
-
-install_rust() {
-    echo "Installing rust..."
-    curl https://sh.rustup.rs -sSf | bash -s -- -y
-    export PATH="$PATH:/root/.cargo/bin"
-}
-
-install_node
-install_rust

ci/manylinux_nodejs/Dockerfile

@@ -1,31 +0,0 @@
-# Many linux dockerfile with Rust, Node, and Lance dependencies installed.
-# This container allows building the node modules native libraries in an
-# environment with a very old glibc, so that we are compatible with a wide
-# range of linux distributions.
-ARG ARCH=x86_64
-
-FROM quay.io/pypa/manylinux2014_${ARCH}
-
-ARG ARCH=x86_64
-ARG DOCKER_USER=default_user
-
-# Install static openssl
-COPY install_openssl.sh install_openssl.sh
-RUN ./install_openssl.sh ${ARCH} > /dev/null
-
-# Protobuf is also installed as root.
-COPY install_protobuf.sh install_protobuf.sh
-RUN ./install_protobuf.sh ${ARCH}
-
-ENV DOCKER_USER=${DOCKER_USER}
-# Create a group and user
-RUN echo ${ARCH} && adduser --user-group --create-home --uid ${DOCKER_USER} build_user
-
-# We switch to the user to install Rust and Node, since those like to be
-# installed at the user level.
-USER ${DOCKER_USER}
-
-COPY prepare_manylinux_node.sh prepare_manylinux_node.sh
-RUN cp /prepare_manylinux_node.sh $HOME/ && \
-    cd $HOME && \
-    ./prepare_manylinux_node.sh ${ARCH}

ci/manylinux_nodejs/build.sh

@@ -1,18 +0,0 @@
-#!/bin/bash
-# Builds the nodejs module for manylinux. Invoked by ci/build_linux_artifacts_nodejs.sh.
-set -e
-ARCH=${1:-x86_64}
-
-if [ "$ARCH" = "x86_64" ]; then
-    export OPENSSL_LIB_DIR=/usr/local/lib64/
-else
-    export OPENSSL_LIB_DIR=/usr/local/lib/
-fi
-export OPENSSL_STATIC=1
-export OPENSSL_INCLUDE_DIR=/usr/local/include/openssl
-
-source $HOME/.bashrc
-
-cd nodejs
-npm ci
-npm run build-release

ci/manylinux_nodejs/install_openssl.sh

@@ -1,26 +0,0 @@
-#!/bin/bash
-# Builds openssl from source so we can statically link to it
-
-# this is to avoid the error we get with the system installation:
-# /usr/bin/ld: <library>: version node not found for symbol SSLeay@@OPENSSL_1.0.1
-# /usr/bin/ld: failed to set dynamic section sizes: Bad value
-set -e
-
-git clone -b OpenSSL_1_1_1u \
-    --single-branch \
-    https://github.com/openssl/openssl.git
-
-pushd openssl
-
-if [[ $1 == x86_64* ]]; then
-    ARCH=linux-x86_64
-else
-    # gnu target
-    ARCH=linux-aarch64
-fi
-
-./Configure no-shared $ARCH
-
-make
-
-make install

ci/manylinux_nodejs/install_protobuf.sh

@@ -1,15 +0,0 @@
-#!/bin/bash
-# Installs protobuf compiler. Should be run as root.
-set -e
-
-if [[ $1 == x86_64* ]]; then
-    ARCH=x86_64
-else
-    # gnu target
-    ARCH=aarch_64
-fi
-
-PB_REL=https://github.com/protocolbuffers/protobuf/releases
-PB_VERSION=23.1
-curl -LO $PB_REL/download/v$PB_VERSION/protoc-$PB_VERSION-linux-$ARCH.zip
-unzip protoc-$PB_VERSION-linux-$ARCH.zip -d /usr/local

ci/manylinux_nodejs/prepare_manylinux_node.sh

@@ -1,21 +0,0 @@
-#!/bin/bash
-set -e
-
-install_node() {
-    echo "Installing node..."
-
-    curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.34.0/install.sh | bash
-
-    source "$HOME"/.bashrc
-
-    nvm install --no-progress 16
-}
-
-install_rust() {
-    echo "Installing rust..."
-    curl https://sh.rustup.rs -sSf | bash -s -- -y
-    export PATH="$PATH:/root/.cargo/bin"
-}
-
-install_node
-install_rust

ci/mock_openai.py

@@ -0,0 +1,57 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright The LanceDB Authors
+"""A zero-dependency mock OpenAI embeddings API endpoint for testing purposes."""
+import argparse
+import json
+import http.server
+
+
+class MockOpenAIRequestHandler(http.server.BaseHTTPRequestHandler):
+    def do_POST(self):
+        content_length = int(self.headers["Content-Length"])
+        post_data = self.rfile.read(content_length)
+        post_data = json.loads(post_data.decode("utf-8"))
+        # See: https://platform.openai.com/docs/api-reference/embeddings/create
+
+        if isinstance(post_data["input"], str):
+            num_inputs = 1
+        else:
+            num_inputs = len(post_data["input"])
+
+        model = post_data.get("model", "text-embedding-ada-002")
+
+        data = []
+        for i in range(num_inputs):
+            data.append({
+                "object": "embedding",
+                "embedding": [0.1] * 1536,
+                "index": i,
+            })
+
+        response = {
+            "object": "list",
+            "data": data,
+            "model": model,
+            "usage": {
+                "prompt_tokens": 0,
+                "total_tokens": 0,
+            }
+        }
+
+        self.send_response(200)
+        self.send_header("Content-type", "application/json")
+        self.end_headers()
+        self.wfile.write(json.dumps(response).encode("utf-8"))
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Mock OpenAI embeddings API endpoint")
+    parser.add_argument("--port", type=int, default=8000, help="Port to listen on")
+    args = parser.parse_args()
+    port = args.port
+
+    print(f"server started on port {port}. Press Ctrl-C to stop.")
+    print(f"To use, set OPENAI_BASE_URL=http://localhost:{port} in your environment.")
+
+    with http.server.HTTPServer(("0.0.0.0", port), MockOpenAIRequestHandler) as server:
+        server.serve_forever()

ci/parse_requirements.py

@@ -0,0 +1,41 @@
+import argparse
+import toml
+
+
+def parse_dependencies(pyproject_path, extras=None):
+    with open(pyproject_path, "r") as file:
+        pyproject = toml.load(file)
+
+    dependencies = pyproject.get("project", {}).get("dependencies", [])
+    for dependency in dependencies:
+        print(dependency)
+
+    optional_dependencies = pyproject.get("project", {}).get(
+        "optional-dependencies", {}
+    )
+
+    if extras:
+        for extra in extras.split(","):
+            for dep in optional_dependencies.get(extra, []):
+                print(dep)
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Generate requirements.txt from pyproject.toml"
+    )
+    parser.add_argument("path", type=str, help="Path to pyproject.toml")
+    parser.add_argument(
+        "--extras",
+        type=str,
+        help="Comma-separated list of extras to include",
+        default="",
+    )
+
+    args = parser.parse_args()
+
+    parse_dependencies(args.path, args.extras)
+
+
+if __name__ == "__main__":
+    main()

ci/set_lance_version.py

@@ -0,0 +1,188 @@
+import argparse
+import sys
+import json
+
+
+def run_command(command: str) -> str:
+    """
+    Run a shell command and return stdout as a string.
+    If exit code is not 0, raise an exception with the stderr output.
+    """
+    import subprocess
+
+    result = subprocess.run(command, shell=True, capture_output=True, text=True)
+    if result.returncode != 0:
+        raise Exception(f"Command failed with error: {result.stderr.strip()}")
+    return result.stdout.strip()
+
+
+def get_latest_stable_version() -> str:
+    version_line = run_command("cargo info lance | grep '^version:'")
+    version = version_line.split(" ")[1].strip()
+    return version
+
+
+def get_latest_preview_version() -> str:
+    lance_tags = run_command(
+        "git ls-remote --tags https://github.com/lancedb/lance.git | grep 'refs/tags/v[0-9beta.-]\\+$'"
+    ).splitlines()
+    lance_tags = (
+        tag.split("refs/tags/")[1]
+        for tag in lance_tags
+        if "refs/tags/" in tag and "beta" in tag
+    )
+    from packaging.version import Version
+
+    latest = max(
+        (tag[1:] for tag in lance_tags if tag.startswith("v")), key=lambda t: Version(t)
+    )
+    return str(latest)
+
+
+def extract_features(line: str) -> list:
+    """
+    Extracts the features from a line in Cargo.toml.
+    Example: 'lance = { "version" = "=0.29.0", "features" = ["dynamodb"] }'
+    Returns: ['dynamodb']
+    """
+    import re
+
+    match = re.search(r'"features"\s*=\s*\[\s*(.*?)\s*\]', line, re.DOTALL)
+    if match:
+        features_str = match.group(1)
+        return [f.strip('"') for f in features_str.split(",") if len(f) > 0]
+    return []
+
+
+def update_cargo_toml(line_updater):
+    """
+    Updates the Cargo.toml file by applying the line_updater function to each line.
+    The line_updater function should take a line as input and return the updated line.
+    """
+    with open("Cargo.toml", "r") as f:
+        lines = f.readlines()
+
+    new_lines = []
+    lance_line = ""
+    is_parsing_lance_line = False
+    for line in lines:
+        if line.startswith("lance"):
+            # Update the line using the provided function
+            if line.strip().endswith("}"):
+                new_lines.append(line_updater(line))
+            else:
+                lance_line = line
+                is_parsing_lance_line = True
+        elif is_parsing_lance_line:
+            lance_line += line
+            if line.strip().endswith("}"):
+                new_lines.append(line_updater(lance_line))
+                lance_line = ""
+                is_parsing_lance_line = False
+            else:
+                print("doesn't end with }:", line)
+        else:
+            # Keep the line unchanged
+            new_lines.append(line)
+
+    with open("Cargo.toml", "w") as f:
+        f.writelines(new_lines)
+
+
+def set_stable_version(version: str):
+    """
+    Sets lines to
+    lance = { "version" = "=0.29.0", "features" = ["dynamodb"] }
+    lance-io = "=0.29.0"
+    ...
+    """
+
+    def line_updater(line: str) -> str:
+        package_name = line.split("=", maxsplit=1)[0].strip()
+        features = extract_features(line)
+        if features:
+            return f'{package_name} = {{ "version" = "={version}", "features" = {json.dumps(features)} }}\n'
+        else:
+            return f'{package_name} = "={version}"\n'
+
+    update_cargo_toml(line_updater)
+
+
+def set_preview_version(version: str):
+    """
+    Sets lines to
+    lance = { "version" = "=0.29.0", "features" = ["dynamodb"], tag = "v0.29.0-beta.2", git="https://github.com/lancedb/lance.git" }
+    lance-io = { version = "=0.29.0", tag = "v0.29.0-beta.2", git="https://github.com/lancedb/lance.git" }
+    ...
+    """
+
+    def line_updater(line: str) -> str:
+        package_name = line.split("=", maxsplit=1)[0].strip()
+        features = extract_features(line)
+        base_version = version.split("-")[0]  # Get the base version without beta suffix
+        if features:
+            return f'{package_name} = {{ "version" = "={base_version}", "features" = {json.dumps(features)}, "tag" = "v{version}", "git" = "https://github.com/lancedb/lance.git" }}\n'
+        else:
+            return f'{package_name} = {{ "version" = "={base_version}", "tag" = "v{version}", "git" = "https://github.com/lancedb/lance.git" }}\n'
+
+    update_cargo_toml(line_updater)
+
+
+def set_local_version():
+    """
+    Sets lines to
+    lance = { path = "../lance/rust/lance", features = ["dynamodb"] }
+    lance-io = { path = "../lance/rust/lance-io" }
+    ...
+    """
+
+    def line_updater(line: str) -> str:
+        package_name = line.split("=", maxsplit=1)[0].strip()
+        features = extract_features(line)
+        if features:
+            return f'{package_name} = {{ "path" = "../lance/rust/{package_name}", "features" = {json.dumps(features)} }}\n'
+        else:
+            return f'{package_name} = {{ "path" = "../lance/rust/{package_name}" }}\n'
+
+    update_cargo_toml(line_updater)
+
+
+parser = argparse.ArgumentParser(description="Set the version of the Lance package.")
+parser.add_argument(
+    "version",
+    type=str,
+    help="The version to set for the Lance package. Use 'stable' for the latest stable version, 'preview' for latest preview version, or a specific version number (e.g., '0.1.0'). You can also specify 'local' to use a local path.",
+)
+args = parser.parse_args()
+
+if args.version == "stable":
+    latest_stable_version = get_latest_stable_version()
+    print(
+        f"Found latest stable version: \033[1mv{latest_stable_version}\033[0m",
+        file=sys.stderr,
+    )
+    set_stable_version(latest_stable_version)
+elif args.version == "preview":
+    latest_preview_version = get_latest_preview_version()
+    print(
+        f"Found latest preview version: \033[1mv{latest_preview_version}\033[0m",
+        file=sys.stderr,
+    )
+    set_preview_version(latest_preview_version)
+elif args.version == "local":
+    set_local_version()
+else:
+    # Parse the version number.
+    version = args.version
+    # Ignore initial v if present.
+    if version.startswith("v"):
+        version = version[1:]
+
+    if "beta" in version:
+        set_preview_version(version)
+    else:
+        set_stable_version(version)
+
+print("Updating lockfiles...", file=sys.stderr, end="")
+run_command("cargo metadata > /dev/null")
+print(" done.", file=sys.stderr)

ci/sysroot-aarch64-pc-windows-msvc.sh

@@ -0,0 +1,105 @@
+#!/bin/sh
+
+# https://github.com/mstorsjo/msvc-wine/blob/master/vsdownload.py
+# https://github.com/mozilla/gecko-dev/blob/6027d1d91f2d3204a3992633b3ef730ff005fc64/build/vs/vs2022-car.yaml
+
+# function dl() {
+# 	curl -O https://download.visualstudio.microsoft.com/download/pr/$1
+# }
+
+# [[.h]]
+
+# "id": "Win11SDK_10.0.26100"
+# "version": "10.0.26100.7"
+
+# libucrt.lib
+
+# example: <assert.h>
+# dir: ucrt/
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/2ee3a5fc6e9fc832af7295b138e93839/universal%20crt%20headers%20libraries%20and%20sources-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/b1aa09b90fe314aceb090f6ec7626624/16ab2ea2187acffa6435e334796c8c89.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/400609bb0ff5804e36dbe6dcd42a7f01/6ee7bbee8435130a869cf971694fd9e2.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/2ac327317abb865a0e3f56b2faefa918/78fa3c824c2c48bd4a49ab5969adaaf7.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/f034bc0b2680f67dccd4bfeea3d0f932/7afc7b670accd8e3cc94cfffd516f5cb.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/7ed5e12f9d50f80825a8b27838cf4c7f/96076045170fe5db6d5dcf14b6f6688e.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/764edc185a696bda9e07df8891dddbbb/a1e2a83aa8a71c48c742eeaff6e71928.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/66854bedc6dbd5ccb5dd82c8e2412231/b2f03f34ff83ec013b9e45c7cd8e8a73.cab
+
+# example: <windows.h>
+# dir: um/
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/b286efac4d83a54fc49190bddef1edc9/windows%20sdk%20for%20windows%20store%20apps%20headers-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/e0dc3811d92ab96fcb72bf63d6c08d71/766c0ffd568bbb31bf7fb6793383e24a.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/613503da4b5628768497822826aed39f/8125ee239710f33ea485965f76fae646.cab
+
+# example: <winapifamily.h>
+# dir: /shared
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/122979f0348d3a2a36b6aa1a111d5d0c/windows%20sdk%20for%20windows%20store%20apps%20headers%20onecoreuap-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/766e04beecdfccff39e91dd9eb32834a/e89e3dcbb016928c7e426238337d69eb.cab
+
+
+# "id": "Microsoft.VisualC.14.16.CRT.Headers"
+# "version": "14.16.27045"
+
+# example: <vcruntime.h>
+# dir: MSVC/
+curl -O https://download.visualstudio.microsoft.com/download/pr/bac0afd7-cc9e-4182-8a83-9898fa20e092/87bbe41e09a2f83711e72696f49681429327eb7a4b90618c35667a6ba2e2880e/Microsoft.VisualC.14.16.CRT.Headers.vsix
+
+# [[.lib]]
+
+# advapi32.lib bcrypt.lib kernel32.lib ntdll.lib user32.lib uuid.lib ws2_32.lib userenv.lib cfgmgr32.lib runtimeobject.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/944c4153b849a1f7d0c0404a4f1c05ea/windows%20sdk%20for%20windows%20store%20apps%20libs-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/5306aed3e1a38d1e8bef5934edeb2a9b/05047a45609f311645eebcac2739fc4c.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/13c8a73a0f5a6474040b26d016a26fab/13d68b8a7b6678a368e2d13ff4027521.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/149578fb3b621cdb61ee1813b9b3e791/463ad1b0783ebda908fd6c16a4abfe93.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/5c986c4f393c6b09d5aec3b539e9fb4a/5a22e5cde814b041749fb271547f4dd5.cab
+
+# dbghelp.lib fwpuclnt.lib arm64rt.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/7a332420d812f7c1d41da865ae5a7c52/windows%20sdk%20desktop%20libs%20arm64-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/19de98ed4a79938d0045d19c047936b3/3e2f7be479e3679d700ce0782e4cc318.cab
+
+# libcmt.lib libvcruntime.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/bac0afd7-cc9e-4182-8a83-9898fa20e092/227f40682a88dc5fa0ccb9cadc9ad30af99ad1f1a75db63407587d079f60d035/Microsoft.VisualC.14.16.CRT.ARM64.Desktop.vsix
+
+
+msiextract universal%20crt%20headers%20libraries%20and%20sources-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20headers-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20headers%20onecoreuap-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20libs-x86_en-us.msi
+msiextract windows%20sdk%20desktop%20libs%20arm64-x86_en-us.msi
+unzip -o Microsoft.VisualC.14.16.CRT.Headers.vsix
+unzip -o Microsoft.VisualC.14.16.CRT.ARM64.Desktop.vsix
+
+mkdir -p /usr/aarch64-pc-windows-msvc/usr/include
+mkdir -p /usr/aarch64-pc-windows-msvc/usr/lib
+
+# lowercase folder/file names
+echo "$(find . -regex ".*/[^/]*[A-Z][^/]*")" | xargs -I{} sh -c 'mv "$(echo "{}" | sed -E '"'"'s/(.*\/)/\L\1/'"'"')" "$(echo "{}" | tr [A-Z] [a-z])"'
+
+# .h
+(cd 'program files/windows kits/10/include/10.0.26100.0' && cp -r ucrt/* um/* shared/* -t /usr/aarch64-pc-windows-msvc/usr/include)
+
+cp -r contents/vc/tools/msvc/14.16.27023/include/* /usr/aarch64-pc-windows-msvc/usr/include
+
+# lowercase #include "" and #include <>
+find /usr/aarch64-pc-windows-msvc/usr/include -type f -exec sed -i -E 's/(#include <[^<>]*?[A-Z][^<>]*?>)|(#include "[^"]*?[A-Z][^"]*?")/\L\1\2/' "{}" ';'
+
+# ARM intrinsics
+# original dir: MSVC/
+
+# '__n128x4' redefined in arm_neon.h
+# "arm64_neon.h" included from intrin.h
+
+(cd /usr/lib/llvm19/lib/clang/19/include && cp arm_neon.h intrin.h -t /usr/aarch64-pc-windows-msvc/usr/include)
+
+# .lib
+
+# _Interlocked intrinsics
+# must always link with arm64rt.lib
+# reason: https://developercommunity.visualstudio.com/t/libucrtlibstreamobj-error-lnk2001-unresolved-exter/1544787#T-ND1599818
+# I don't understand the 'correct' fix for this, arm64rt.lib is supposed to be the workaround
+
+(cd 'program files/windows kits/10/lib/10.0.26100.0/um/arm64' && cp advapi32.lib bcrypt.lib kernel32.lib ntdll.lib user32.lib uuid.lib ws2_32.lib userenv.lib cfgmgr32.lib runtimeobject.lib dbghelp.lib fwpuclnt.lib arm64rt.lib -t /usr/aarch64-pc-windows-msvc/usr/lib)
+
+(cd 'contents/vc/tools/msvc/14.16.27023/lib/arm64' && cp libcmt.lib libvcruntime.lib -t /usr/aarch64-pc-windows-msvc/usr/lib)
+
+cp 'program files/windows kits/10/lib/10.0.26100.0/ucrt/arm64/libucrt.lib' /usr/aarch64-pc-windows-msvc/usr/lib

ci/sysroot-x86_64-pc-windows-msvc.sh

@@ -0,0 +1,105 @@
+#!/bin/sh
+
+# https://github.com/mstorsjo/msvc-wine/blob/master/vsdownload.py
+# https://github.com/mozilla/gecko-dev/blob/6027d1d91f2d3204a3992633b3ef730ff005fc64/build/vs/vs2022-car.yaml
+
+# function dl() {
+# 	curl -O https://download.visualstudio.microsoft.com/download/pr/$1
+# }
+
+# [[.h]]
+
+# "id": "Win11SDK_10.0.26100"
+# "version": "10.0.26100.7"
+
+# libucrt.lib
+
+# example: <assert.h>
+# dir: ucrt/
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/2ee3a5fc6e9fc832af7295b138e93839/universal%20crt%20headers%20libraries%20and%20sources-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/b1aa09b90fe314aceb090f6ec7626624/16ab2ea2187acffa6435e334796c8c89.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/400609bb0ff5804e36dbe6dcd42a7f01/6ee7bbee8435130a869cf971694fd9e2.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/2ac327317abb865a0e3f56b2faefa918/78fa3c824c2c48bd4a49ab5969adaaf7.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/f034bc0b2680f67dccd4bfeea3d0f932/7afc7b670accd8e3cc94cfffd516f5cb.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/7ed5e12f9d50f80825a8b27838cf4c7f/96076045170fe5db6d5dcf14b6f6688e.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/764edc185a696bda9e07df8891dddbbb/a1e2a83aa8a71c48c742eeaff6e71928.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/66854bedc6dbd5ccb5dd82c8e2412231/b2f03f34ff83ec013b9e45c7cd8e8a73.cab
+
+# example: <windows.h>
+# dir: um/
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/b286efac4d83a54fc49190bddef1edc9/windows%20sdk%20for%20windows%20store%20apps%20headers-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/e0dc3811d92ab96fcb72bf63d6c08d71/766c0ffd568bbb31bf7fb6793383e24a.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/613503da4b5628768497822826aed39f/8125ee239710f33ea485965f76fae646.cab
+
+# example: <winapifamily.h>
+# dir: /shared
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/122979f0348d3a2a36b6aa1a111d5d0c/windows%20sdk%20for%20windows%20store%20apps%20headers%20onecoreuap-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/766e04beecdfccff39e91dd9eb32834a/e89e3dcbb016928c7e426238337d69eb.cab
+
+
+# "id": "Microsoft.VisualC.14.16.CRT.Headers"
+# "version": "14.16.27045"
+
+# example: <vcruntime.h>
+# dir: MSVC/
+curl -O https://download.visualstudio.microsoft.com/download/pr/bac0afd7-cc9e-4182-8a83-9898fa20e092/87bbe41e09a2f83711e72696f49681429327eb7a4b90618c35667a6ba2e2880e/Microsoft.VisualC.14.16.CRT.Headers.vsix
+
+# [[.lib]]
+
+# advapi32.lib bcrypt.lib kernel32.lib ntdll.lib user32.lib uuid.lib ws2_32.lib userenv.lib cfgmgr32.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/944c4153b849a1f7d0c0404a4f1c05ea/windows%20sdk%20for%20windows%20store%20apps%20libs-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/5306aed3e1a38d1e8bef5934edeb2a9b/05047a45609f311645eebcac2739fc4c.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/13c8a73a0f5a6474040b26d016a26fab/13d68b8a7b6678a368e2d13ff4027521.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/149578fb3b621cdb61ee1813b9b3e791/463ad1b0783ebda908fd6c16a4abfe93.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/5c986c4f393c6b09d5aec3b539e9fb4a/5a22e5cde814b041749fb271547f4dd5.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/bfc3904a0195453419ae4dfea7abd6fb/e10768bb6e9d0ea730280336b697da66.cab
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/637f9f3be880c71f9e3ca07b4d67345c/f9b24c8280986c0683fbceca5326d806.cab
+
+# dbghelp.lib fwpuclnt.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/9f51690d5aa804b1340ce12d1ec80f89/windows%20sdk%20desktop%20libs%20x64-x86_en-us.msi
+curl -O https://download.visualstudio.microsoft.com/download/pr/32863b8d-a46d-4231-8e84-0888519d20a9/d3a7df4ca3303a698640a29e558a5e5b/58314d0646d7e1a25e97c902166c3155.cab
+
+# libcmt.lib libvcruntime.lib
+curl -O https://download.visualstudio.microsoft.com/download/pr/bac0afd7-cc9e-4182-8a83-9898fa20e092/8728f21ae09940f1f4b4ee47b4a596be2509e2a47d2f0c83bbec0ea37d69644b/Microsoft.VisualC.14.16.CRT.x64.Desktop.vsix
+
+
+msiextract universal%20crt%20headers%20libraries%20and%20sources-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20headers-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20headers%20onecoreuap-x86_en-us.msi
+msiextract windows%20sdk%20for%20windows%20store%20apps%20libs-x86_en-us.msi
+msiextract windows%20sdk%20desktop%20libs%20x64-x86_en-us.msi
+unzip -o Microsoft.VisualC.14.16.CRT.Headers.vsix
+unzip -o Microsoft.VisualC.14.16.CRT.x64.Desktop.vsix
+
+mkdir -p /usr/x86_64-pc-windows-msvc/usr/include
+mkdir -p /usr/x86_64-pc-windows-msvc/usr/lib
+
+# lowercase folder/file names
+echo "$(find . -regex ".*/[^/]*[A-Z][^/]*")" | xargs -I{} sh -c 'mv "$(echo "{}" | sed -E '"'"'s/(.*\/)/\L\1/'"'"')" "$(echo "{}" | tr [A-Z] [a-z])"'
+
+# .h
+(cd 'program files/windows kits/10/include/10.0.26100.0' && cp -r ucrt/* um/* shared/* -t /usr/x86_64-pc-windows-msvc/usr/include)
+
+cp -r contents/vc/tools/msvc/14.16.27023/include/* /usr/x86_64-pc-windows-msvc/usr/include
+
+# lowercase #include "" and #include <>
+find /usr/x86_64-pc-windows-msvc/usr/include -type f -exec sed -i -E 's/(#include <[^<>]*?[A-Z][^<>]*?>)|(#include "[^"]*?[A-Z][^"]*?")/\L\1\2/' "{}" ';'
+
+# x86 intrinsics
+# original dir: MSVC/
+
+# '_mm_movemask_epi8' defined in emmintrin.h
+# '__v4sf' defined in xmmintrin.h
+# '__v2si' defined in mmintrin.h
+# '__m128d' redefined in immintrin.h
+# '__m128i' redefined in intrin.h
+# '_mm_comlt_epu8' defined in ammintrin.h
+
+(cd /usr/lib/llvm19/lib/clang/19/include && cp emmintrin.h xmmintrin.h mmintrin.h immintrin.h intrin.h ammintrin.h -t /usr/x86_64-pc-windows-msvc/usr/include)
+
+# .lib
+(cd 'program files/windows kits/10/lib/10.0.26100.0/um/x64' && cp advapi32.lib bcrypt.lib kernel32.lib ntdll.lib user32.lib uuid.lib ws2_32.lib userenv.lib cfgmgr32.lib dbghelp.lib fwpuclnt.lib -t /usr/x86_64-pc-windows-msvc/usr/lib)
+
+(cd 'contents/vc/tools/msvc/14.16.27023/lib/x64' && cp libcmt.lib libvcruntime.lib -t /usr/x86_64-pc-windows-msvc/usr/lib)
+
+cp 'program files/windows kits/10/lib/10.0.26100.0/ucrt/x64/libucrt.lib' /usr/x86_64-pc-windows-msvc/usr/lib

ci/update_lockfiles.sh

@@ -0,0 +1,27 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+AMEND=false
+
+for arg in "$@"; do
+  if [[ "$arg" == "--amend" ]]; then
+    AMEND=true
+  fi
+done
+
+# This updates the lockfile without building
+cargo metadata --quiet > /dev/null
+
+pushd nodejs || exit 1
+npm install --package-lock-only --silent
+popd
+
+if git diff --quiet --exit-code; then
+  echo "No lockfile changes to commit; skipping amend."
+elif $AMEND; then
+  git add Cargo.lock nodejs/package-lock.json
+  git commit --amend --no-edit
+else
+  git add Cargo.lock nodejs/package-lock.json
+  git commit -m "Update lockfiles"
+fi

ci/validate_stable_lance.py

@@ -0,0 +1,34 @@
+import tomllib
+
+found_preview_lance = False
+
+with open("Cargo.toml", "rb") as f:
+    cargo_data = tomllib.load(f)
+
+    for name, dep in cargo_data["workspace"]["dependencies"].items():
+        if name == "lance" or name.startswith("lance-"):
+            if isinstance(dep, str):
+                version = dep
+            elif isinstance(dep, dict):
+                # Version doesn't have the beta tag in it, so we instead look
+                # at the git tag.
+                version = dep.get('tag', dep.get('version'))
+            else:
+                raise ValueError("Unexpected type for dependency: " + str(dep))
+
+            if "beta" in version:
+                found_preview_lance = True
+                print(f"Dependency '{name}' is a preview version: {version}")
+
+with open("python/pyproject.toml", "rb") as f:
+    py_proj_data = tomllib.load(f)
+
+    for dep in py_proj_data["project"]["dependencies"]:
+        if dep.startswith("pylance"):
+            if "b" in dep:
+                found_preview_lance = True
+                print(f"Dependency '{dep}' is a preview version")
+            break  # Only one pylance dependency
+
+if found_preview_lance:
+    raise ValueError("Found preview version of Lance in dependencies")

docs/README.md

@@ -2,43 +2,88 @@
 
 LanceDB docs are deployed to https://lancedb.github.io/lancedb/.
 
-Docs is built and deployed automatically by [Github Actions](.github/workflows/docs.yml)
+Docs is built and deployed automatically by [Github Actions](../.github/workflows/docs.yml)
 whenever a commit is pushed to the `main` branch. So it is possible for the docs to show
 unreleased features.
 
 ## Building the docs
 
 ### Setup
-1. Install LanceDB. From LanceDB repo root: `pip install -e python`
-2. Install dependencies. From LanceDB repo root: `pip install -r docs/requirements.txt`
-3. Make sure you have node and npm setup
-4. Make sure protobuf and libssl are installed
+1. Install LanceDB Python. See setup in [Python contributing guide](../python/CONTRIBUTING.md).
+   Run `make develop` to install the Python package.
+2. Install documentation dependencies. From LanceDB repo root: `pip install -r docs/requirements.txt`
 
-### Building node module and create markdown files
+### Preview the docs
 
-See [Javascript docs README](./src/javascript/README.md)
-
-### Build docs
-From LanceDB repo root:
-
-Run: `PYTHONPATH=. mkdocs build -f docs/mkdocs.yml`
-
-If successful, you should see a `docs/site` directory that you can verify locally.
-
-### Run local server
-
-You can run a local server to test the docs prior to deployment by navigating to the `docs` directory and running the following command:
-
-```bash
+```shell
 cd docs
 mkdocs serve
 ```
 
-### Run doctest for typescript example
+If you want to just generate the HTML files:
 
-```bash
-cd lancedb/docs
-npm i
-npm run build
-npm run all
+```shell
+PYTHONPATH=. mkdocs build -f docs/mkdocs.yml
+```
+
+If successful, you should see a `docs/site` directory that you can verify locally.
+
+## Adding examples
+
+To make sure examples are correct, we put examples in test files so they can be
+run as part of our test suites.
+
+You can see the tests are at:
+
+* Python: `python/python/tests/docs`
+* Typescript: `nodejs/examples/`
+
+### Checking python examples
+
+```shell
+cd python
+pytest -vv python/tests/docs
+```
+
+### Checking typescript examples
+
+The `@lancedb/lancedb` package must be built before running the tests:
+
+```shell
+pushd nodejs
+npm ci
+npm run build
+popd
+```
+
+Then you can run the examples by going to the `nodejs/examples` directory and
+running the tests like a normal npm package:
+
+```shell
+pushd nodejs/examples
+npm ci
+npm test
+popd
+```
+
+## API documentation
+
+### Python
+
+The Python API documentation is organized based on the file `docs/src/python/python.md`.
+We manually add entries there so we can control the organization of the reference page.
+**However, this means any new types must be manually added to the file.** No additional
+steps are needed to generate the API documentation.
+
+### Typescript
+
+The typescript API documentation is generated from the typescript source code using [typedoc](https://typedoc.org/).
+
+When new APIs are added, you must manually re-run the typedoc command to update the API documentation.
+The new files should be checked into the repository.
+
+```shell
+pushd nodejs
+npm run docs
+popd
 ```

docs/mkdocs.yml

@@ -4,6 +4,9 @@ repo_url: https://github.com/lancedb/lancedb
 edit_uri: https://github.com/lancedb/lancedb/tree/main/docs/src
 repo_name: lancedb/lancedb
 docs_dir: src
+watch:
+  - src
+  - ../python/python
 
 theme:
   name: "material"
@@ -26,6 +29,7 @@ theme:
     - content.code.copy
     - content.tabs.link
     - content.action.edit
+    - content.tooltips
     - toc.follow
     - navigation.top
     - navigation.tabs
@@ -33,8 +37,10 @@ theme:
     - navigation.footer
     - navigation.tracking
     - navigation.instant
+    - content.footnote.tooltips
   icon:
     repo: fontawesome/brands/github
+    annotation: material/arrow-right-circle
   custom_dir: overrides
 
 plugins:
@@ -52,17 +58,27 @@ plugins:
             show_signature_annotations: true
             show_root_heading: true
             members_order: source
+            docstring_section_style: list
+            signature_crossrefs: true
+            separate_signature: true
           import:
             # for cross references
             - https://arrow.apache.org/docs/objects.inv
             - https://pandas.pydata.org/docs/objects.inv
+            - https://lancedb.github.io/lance/objects.inv
+            - https://docs.pydantic.dev/latest/objects.inv
   - mkdocs-jupyter
   - render_swagger:
-      allow_arbitrary_locations : true
+      allow_arbitrary_locations: true
 
 markdown_extensions:
   - admonition
   - footnotes
+  - pymdownx.critic
+  - pymdownx.caret
+  - pymdownx.keys
+  - pymdownx.mark
+  - pymdownx.tilde
   - pymdownx.details
   - pymdownx.highlight:
       anchor_linenums: true
@@ -76,7 +92,15 @@ markdown_extensions:
   - pymdownx.tabbed:
       alternate_style: true
   - md_in_html
+  - abbr
   - attr_list
+  - pymdownx.snippets
+  - pymdownx.emoji:
+      emoji_index: !!python/name:material.extensions.emoji.twemoji
+      emoji_generator: !!python/name:material.extensions.emoji.to_svg
+  - markdown.extensions.toc:
+      baselevel: 1
+      permalink: ""
 
 nav:
   - Home:
@@ -84,30 +108,54 @@ nav:
       - 🏃🏼‍♂️ Quick start: basic.md
       - 📚 Concepts:
           - Vector search: concepts/vector_search.md
-          - Indexing: concepts/index_ivfpq.md
+          - Indexing:
+              - IVFPQ: concepts/index_ivfpq.md
+              - HNSW: concepts/index_hnsw.md
           - Storage: concepts/storage.md
           - Data management: concepts/data_management.md
       - 🔨 Guides:
           - Working with tables: guides/tables.md
-          - Building an ANN index: ann_indexes.md
+          - Building a vector index: ann_indexes.md
           - Vector Search: search.md
-          - Full-text search: fts.md
+          - Full-text search (native): fts.md
+          - Full-text search (tantivy-based): fts_tantivy.md
+          - Building a scalar index: guides/scalar_index.md
           - Hybrid search:
               - Overview: hybrid_search/hybrid_search.md
               - Comparing Rerankers: hybrid_search/eval.md
               - Airbnb financial data example: notebooks/hybrid_search.ipynb
+          - Late interaction with MultiVector search:
+              - Overview: guides/multi-vector.md
+              - Example: notebooks/Multivector_on_LanceDB.ipynb
+          - RAG:
+              - Vanilla RAG: rag/vanilla_rag.md
+              - Multi-head RAG: rag/multi_head_rag.md
+              - Corrective RAG: rag/corrective_rag.md
+              - Agentic RAG: rag/agentic_rag.md
+              - Graph RAG: rag/graph_rag.md
+              - Self RAG: rag/self_rag.md
+              - Adaptive RAG: rag/adaptive_rag.md
+              - SFR RAG: rag/sfr_rag.md
+              - Advanced Techniques:
+                  - HyDE: rag/advanced_techniques/hyde.md
+                  - FLARE: rag/advanced_techniques/flare.md
           - Reranking:
               - Quickstart: reranking/index.md
               - Cohere Reranker: reranking/cohere.md
               - Linear Combination Reranker: reranking/linear_combination.md
+              - Reciprocal Rank Fusion Reranker: reranking/rrf.md
               - Cross Encoder Reranker: reranking/cross_encoder.md
               - ColBERT Reranker: reranking/colbert.md
               - Jina Reranker: reranking/jina.md
               - OpenAI Reranker: reranking/openai.md
+              - AnswerDotAi Rerankers: reranking/answerdotai.md
+              - Voyage AI Rerankers: reranking/voyageai.md
               - Building Custom Rerankers: reranking/custom_reranker.md
               - Example: notebooks/lancedb_reranking.ipynb
           - Filtering: sql.md
-          - Versioning & Reproducibility: notebooks/reproducibility.ipynb
+          - Versioning & Reproducibility:
+              - sync API: notebooks/reproducibility.ipynb
+              - async API: notebooks/reproducibility_async.ipynb
           - Configuring Storage: guides/storage.md
           - Migration Guide: migration.md
           - Tuning retrieval performance:
@@ -115,10 +163,29 @@ nav:
               - Reranking: guides/tuning_retrievers/2_reranking.md
               - Embedding fine-tuning: guides/tuning_retrievers/3_embed_tuning.md
       - 🧬 Managing embeddings:
-          - Overview: embeddings/index.md
+          - Understand Embeddings: embeddings/understanding_embeddings.md
+          - Get Started: embeddings/index.md
           - Embedding functions: embeddings/embedding_functions.md
-          - Available models: embeddings/default_embedding_functions.md
+          - Available models:
+              - Overview: embeddings/default_embedding_functions.md
+              - Text Embedding Functions:
+                  - Sentence Transformers: embeddings/available_embedding_models/text_embedding_functions/sentence_transformers.md
+                  - Huggingface Embedding Models: embeddings/available_embedding_models/text_embedding_functions/huggingface_embedding.md
+                  - Ollama Embeddings: embeddings/available_embedding_models/text_embedding_functions/ollama_embedding.md
+                  - OpenAI Embeddings: embeddings/available_embedding_models/text_embedding_functions/openai_embedding.md
+                  - Instructor Embeddings: embeddings/available_embedding_models/text_embedding_functions/instructor_embedding.md
+                  - Gemini Embeddings: embeddings/available_embedding_models/text_embedding_functions/gemini_embedding.md
+                  - Cohere Embeddings: embeddings/available_embedding_models/text_embedding_functions/cohere_embedding.md
+                  - Jina Embeddings: embeddings/available_embedding_models/text_embedding_functions/jina_embedding.md
+                  - AWS Bedrock Text Embedding Functions: embeddings/available_embedding_models/text_embedding_functions/aws_bedrock_embedding.md
+                  - IBM watsonx.ai Embeddings: embeddings/available_embedding_models/text_embedding_functions/ibm_watsonx_ai_embedding.md
+                  - Voyage AI Embeddings: embeddings/available_embedding_models/text_embedding_functions/voyageai_embedding.md
+              - Multimodal Embedding Functions:
+                  - OpenClip embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/openclip_embedding.md
+                  - Imagebind embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/imagebind_embedding.md
+                  - Jina Embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/jina_multimodal_embedding.md
           - User-defined embedding functions: embeddings/custom_embedding_function.md
+          - Variables and secrets: embeddings/variables_and_secrets.md
           - "Example: Multi-lingual semantic search": notebooks/multi_lingual_example.ipynb
           - "Example: MultiModal CLIP Embeddings": notebooks/DisappearingEmbeddingFunction.ipynb
       - 🔌 Integrations:
@@ -126,26 +193,35 @@ nav:
           - Pandas and PyArrow: python/pandas_and_pyarrow.md
           - Polars: python/polars_arrow.md
           - DuckDB: python/duckdb.md
+          - Datafusion: python/datafusion.md
           - LangChain:
-            - LangChain 🔗: integrations/langchain.md
-            - LangChain demo: notebooks/langchain_demo.ipynb
-            - LangChain JS/TS 🔗: https://js.langchain.com/docs/integrations/vectorstores/lancedb
+              - LangChain 🔗: integrations/langchain.md
+              - LangChain demo: notebooks/langchain_demo.ipynb
+              - LangChain JS/TS 🔗: https://js.langchain.com/docs/integrations/vectorstores/lancedb
           - LlamaIndex 🦙:
-            - LlamaIndex docs: integrations/llamaIndex.md
-            - LlamaIndex demo: notebooks/llamaIndex_demo.ipynb
+              - LlamaIndex docs: integrations/llamaIndex.md
+              - LlamaIndex demo: notebooks/llamaIndex_demo.ipynb
           - Pydantic: python/pydantic.md
           - Voxel51: integrations/voxel51.md
           - PromptTools: integrations/prompttools.md
+          - dlt: integrations/dlt.md
+          - phidata: integrations/phidata.md
+          - Genkit: integrations/genkit.md
       - 🎯 Examples:
           - Overview: examples/index.md
           - 🐍 Python:
               - Overview: examples/examples_python.md
-              - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
-              - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
-              - Multimodal search using CLIP: notebooks/multimodal_search.ipynb
-              - Example - Calculate CLIP Embeddings with Roboflow Inference: examples/image_embeddings_roboflow.md
-              - Serverless QA Bot with S3 and Lambda: examples/serverless_lancedb_with_s3_and_lambda.md
-              - Serverless QA Bot with Modal: examples/serverless_qa_bot_with_modal_and_langchain.md
+              - Build From Scratch: examples/python_examples/build_from_scratch.md
+              - Multimodal: examples/python_examples/multimodal.md
+              - Rag: examples/python_examples/rag.md
+              - Vector Search: examples/python_examples/vector_search.md
+              - Chatbot: examples/python_examples/chatbot.md
+              - Evaluation: examples/python_examples/evaluations.md
+              - AI Agent: examples/python_examples/aiagent.md
+              - Recommender System: examples/python_examples/recommendersystem.md
+              - Miscellaneous:
+                  - Serverless QA Bot with S3 and Lambda: examples/serverless_lancedb_with_s3_and_lambda.md
+                  - Serverless QA Bot with Modal: examples/serverless_qa_bot_with_modal_and_langchain.md
           - 👾 JavaScript:
               - Overview: examples/examples_js.md
               - Serverless Website Chatbot: examples/serverless_website_chatbot.md
@@ -153,46 +229,67 @@ nav:
               - TransformersJS Embedding Search: examples/transformerjs_embedding_search_nodejs.md
           - 🦀 Rust:
               - Overview: examples/examples_rust.md
+      - 📓 Studies:
+          - ↗Improve retrievers with hybrid search and reranking: https://blog.lancedb.com/hybrid-search-and-reranking-report/
       - 💭 FAQs: faq.md
+      - 🔍 Troubleshooting: troubleshooting.md
       - ⚙️ API reference:
           - 🐍 Python: python/python.md
           - 👾 JavaScript (vectordb): javascript/modules.md
-          - 👾 JavaScript (lancedb): javascript/modules.md
+          - 👾 JavaScript (lancedb): js/globals.md
           - 🦀 Rust: https://docs.rs/lancedb/latest/lancedb/
-      - ☁️ LanceDB Cloud:
-          - Overview: cloud/index.md
-          - API reference:
-              - 🐍 Python: python/saas-python.md
-              - 👾 JavaScript: javascript/modules.md
-              - REST API: cloud/rest.md
 
   - Quick start: basic.md
   - Concepts:
       - Vector search: concepts/vector_search.md
-      - Indexing: concepts/index_ivfpq.md
+      - Indexing:
+          - IVFPQ: concepts/index_ivfpq.md
+          - HNSW: concepts/index_hnsw.md
       - Storage: concepts/storage.md
       - Data management: concepts/data_management.md
   - Guides:
       - Working with tables: guides/tables.md
+      - Working with SQL: guides/sql_querying.md
       - Building an ANN index: ann_indexes.md
       - Vector Search: search.md
-      - Full-text search: fts.md
+      - Full-text search (native): fts.md
+      - Full-text search (tantivy-based): fts_tantivy.md
+      - Building a scalar index: guides/scalar_index.md
       - Hybrid search:
           - Overview: hybrid_search/hybrid_search.md
           - Comparing Rerankers: hybrid_search/eval.md
           - Airbnb financial data example: notebooks/hybrid_search.ipynb
+      - Late interaction with MultiVector search:
+          - Overview: guides/multi-vector.md
+          - Document search Example: notebooks/Multivector_on_LanceDB.ipynb
+      - RAG:
+          - Vanilla RAG: rag/vanilla_rag.md
+          - Multi-head RAG: rag/multi_head_rag.md
+          - Corrective RAG: rag/corrective_rag.md
+          - Agentic RAG: rag/agentic_rag.md
+          - Graph RAG: rag/graph_rag.md
+          - Self RAG: rag/self_rag.md
+          - Adaptive RAG: rag/adaptive_rag.md
+          - SFR RAG: rag/sfr_rag.md
+          - Advanced Techniques:
+              - HyDE: rag/advanced_techniques/hyde.md
+              - FLARE: rag/advanced_techniques/flare.md
       - Reranking:
           - Quickstart: reranking/index.md
           - Cohere Reranker: reranking/cohere.md
           - Linear Combination Reranker: reranking/linear_combination.md
+          - Reciprocal Rank Fusion Reranker: reranking/rrf.md
           - Cross Encoder Reranker: reranking/cross_encoder.md
           - ColBERT Reranker: reranking/colbert.md
           - Jina Reranker: reranking/jina.md
           - OpenAI Reranker: reranking/openai.md
+          - AnswerDotAi Rerankers: reranking/answerdotai.md
           - Building Custom Rerankers: reranking/custom_reranker.md
           - Example: notebooks/lancedb_reranking.ipynb
       - Filtering: sql.md
-      - Versioning & Reproducibility: notebooks/reproducibility.ipynb
+      - Versioning & Reproducibility:
+          - sync API: notebooks/reproducibility.ipynb
+          - async API: notebooks/reproducibility_async.ipynb
       - Configuring Storage: guides/storage.md
       - Migration Guide: migration.md
       - Tuning retrieval performance:
@@ -200,10 +297,28 @@ nav:
           - Reranking: guides/tuning_retrievers/2_reranking.md
           - Embedding fine-tuning: guides/tuning_retrievers/3_embed_tuning.md
   - Managing Embeddings:
-      - Overview: embeddings/index.md
+      - Understand Embeddings: embeddings/understanding_embeddings.md
+      - Get Started: embeddings/index.md
       - Embedding functions: embeddings/embedding_functions.md
-      - Available models: embeddings/default_embedding_functions.md
+      - Available models:
+          - Overview: embeddings/default_embedding_functions.md
+          - Text Embedding Functions:
+              - Sentence Transformers: embeddings/available_embedding_models/text_embedding_functions/sentence_transformers.md
+              - Huggingface Embedding Models: embeddings/available_embedding_models/text_embedding_functions/huggingface_embedding.md
+              - Ollama Embeddings: embeddings/available_embedding_models/text_embedding_functions/ollama_embedding.md
+              - OpenAI Embeddings: embeddings/available_embedding_models/text_embedding_functions/openai_embedding.md
+              - Instructor Embeddings: embeddings/available_embedding_models/text_embedding_functions/instructor_embedding.md
+              - Gemini Embeddings: embeddings/available_embedding_models/text_embedding_functions/gemini_embedding.md
+              - Cohere Embeddings: embeddings/available_embedding_models/text_embedding_functions/cohere_embedding.md
+              - Jina Embeddings: embeddings/available_embedding_models/text_embedding_functions/jina_embedding.md
+              - AWS Bedrock Text Embedding Functions: embeddings/available_embedding_models/text_embedding_functions/aws_bedrock_embedding.md
+              - IBM watsonx.ai Embeddings: embeddings/available_embedding_models/text_embedding_functions/ibm_watsonx_ai_embedding.md
+          - Multimodal Embedding Functions:
+              - OpenClip embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/openclip_embedding.md
+              - Imagebind embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/imagebind_embedding.md
+              - Jina Embeddings: embeddings/available_embedding_models/multimodal_embedding_functions/jina_multimodal_embedding.md
       - User-defined embedding functions: embeddings/custom_embedding_function.md
+      - Variables and secrets: embeddings/variables_and_secrets.md
       - "Example: Multi-lingual semantic search": notebooks/multi_lingual_example.ipynb
       - "Example: MultiModal CLIP Embeddings": notebooks/DisappearingEmbeddingFunction.ipynb
   - Integrations:
@@ -211,34 +326,47 @@ nav:
       - Pandas and PyArrow: python/pandas_and_pyarrow.md
       - Polars: python/polars_arrow.md
       - DuckDB: python/duckdb.md
+      - Datafusion: python/datafusion.md
       - LangChain 🦜️🔗↗: integrations/langchain.md
       - LangChain.js 🦜️🔗↗: https://js.langchain.com/docs/integrations/vectorstores/lancedb
       - LlamaIndex 🦙↗: integrations/llamaIndex.md
       - Pydantic: python/pydantic.md
       - Voxel51: integrations/voxel51.md
       - PromptTools: integrations/prompttools.md
+      - dlt: integrations/dlt.md
+      - phidata: integrations/phidata.md
+      - Genkit: integrations/genkit.md
   - Examples:
       - examples/index.md
-      - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
-      - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
-      - Multimodal search using CLIP: notebooks/multimodal_search.ipynb
-      - Serverless QA Bot with S3 and Lambda: examples/serverless_lancedb_with_s3_and_lambda.md
-      - Serverless QA Bot with Modal: examples/serverless_qa_bot_with_modal_and_langchain.md
-      - YouTube Transcript Search (JS): examples/youtube_transcript_bot_with_nodejs.md
-      - Serverless Chatbot from any website: examples/serverless_website_chatbot.md
-      - TransformersJS Embedding Search: examples/transformerjs_embedding_search_nodejs.md
+      - 🐍 Python:
+          - Overview: examples/examples_python.md
+          - Build From Scratch: examples/python_examples/build_from_scratch.md
+          - Multimodal: examples/python_examples/multimodal.md
+          - Rag: examples/python_examples/rag.md
+          - Vector Search: examples/python_examples/vector_search.md
+          - Chatbot: examples/python_examples/chatbot.md
+          - Evaluation: examples/python_examples/evaluations.md
+          - AI Agent: examples/python_examples/aiagent.md
+          - Recommender System: examples/python_examples/recommendersystem.md
+          - Miscellaneous:
+              - Serverless QA Bot with S3 and Lambda: examples/serverless_lancedb_with_s3_and_lambda.md
+              - Serverless QA Bot with Modal: examples/serverless_qa_bot_with_modal_and_langchain.md
+      - 👾 JavaScript:
+          - Overview: examples/examples_js.md
+          - Serverless Website Chatbot: examples/serverless_website_chatbot.md
+          - YouTube Transcript Search: examples/youtube_transcript_bot_with_nodejs.md
+          - TransformersJS Embedding Search: examples/transformerjs_embedding_search_nodejs.md
+      - 🦀 Rust:
+          - Overview: examples/examples_rust.md
+  - Studies:
+      - studies/overview.md
+      - ↗Improve retrievers with hybrid search and reranking: https://blog.lancedb.com/hybrid-search-and-reranking-report/
   - API reference:
       - Overview: api_reference.md
       - Python: python/python.md
       - Javascript (vectordb): javascript/modules.md
-      - Javascript (lancedb): js/modules.md
+      - Javascript (lancedb): js/globals.md
       - Rust: https://docs.rs/lancedb/latest/lancedb/index.html
-  - LanceDB Cloud:
-      - Overview: cloud/index.md
-      - API reference:
-          - 🐍 Python: python/saas-python.md
-          - 👾 JavaScript: javascript/modules.md
-          - REST API: cloud/rest.md
 
 extra_css:
   - styles/global.css
@@ -246,6 +374,7 @@ extra_css:
 
 extra_javascript:
   - "extra_js/init_ask_ai_widget.js"
+  - "extra_js/reo.js"
 
 extra:
   analytics:

docs/openapi.yml

@@ -38,6 +38,13 @@ components:
       required: true
       schema:
         type: string
+    index_name:
+      name: index_name
+      in: path
+      description: name of the index
+      required: true
+      schema:
+        type: string
   responses:
     invalid_request:
       description: Invalid request
@@ -164,7 +171,7 @@ paths:
                 distance_type:
                   type: string
                   description: |
-                    The distance metric to use for search. L2, Cosine, Dot and Hamming are supported. Default is L2.
+                    The distance metric to use for search. l2, Cosine, Dot and Hamming are supported. Default is l2.
                 bypass_vector_index:
                   type: boolean
                   description: |
@@ -443,7 +450,7 @@ paths:
                   type: string
                   nullable: false
                   description: |
-                    The metric type to use for the index. L2, Cosine, Dot are supported.
+                    The metric type to use for the index. l2, Cosine, Dot are supported.
                 index_type:
                   type: string
       responses:
@@ -485,3 +492,22 @@ paths:
           $ref: "#/components/responses/unauthorized"
         "404":
           $ref: "#/components/responses/not_found"
+  /v1/table/{name}/index/{index_name}/drop/:
+    post:
+      description: Drop an index from the table
+      tags:
+        - Tables
+      summary: Drop an index from the table
+      operationId: dropIndex
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+        - $ref: "#/components/parameters/index_name"
+      responses:
+        "200":
+          description: Index successfully dropped
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"

docs/overrides/partials/main.html

@@ -0,0 +1,5 @@
+{% extends "base.html" %}
+
+{% block announce %}
+  📚 Starting June 1st, 2025, please use <a href="https://lancedb.github.io/documentation" target="_blank" rel="noopener noreferrer">lancedb.github.io/documentation</a> for the latest docs.
+{% endblock %}

docs/package-lock.json

@@ -19,7 +19,7 @@
     },
     "../node": {
       "name": "vectordb",
-      "version": "0.4.6",
+      "version": "0.21.2-beta.0",
       "cpu": [
         "x64",
         "arm64"
@@ -31,9 +31,7 @@
         "win32"
       ],
       "dependencies": {
-        "@apache-arrow/ts": "^14.0.2",
         "@neon-rs/load": "^0.0.74",
-        "apache-arrow": "^14.0.2",
         "axios": "^1.4.0"
       },
       "devDependencies": {
@@ -46,6 +44,7 @@
         "@types/temp": "^0.9.1",
         "@types/uuid": "^9.0.3",
         "@typescript-eslint/eslint-plugin": "^5.59.1",
+        "apache-arrow-old": "npm:apache-arrow@13.0.0",
         "cargo-cp-artifact": "^0.1",
         "chai": "^4.3.7",
         "chai-as-promised": "^7.1.1",
@@ -62,15 +61,19 @@
         "ts-node-dev": "^2.0.0",
         "typedoc": "^0.24.7",
         "typedoc-plugin-markdown": "^3.15.3",
-        "typescript": "*",
+        "typescript": "^5.1.0",
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.4.6",
-        "@lancedb/vectordb-darwin-x64": "0.4.6",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.4.6",
-        "@lancedb/vectordb-linux-x64-gnu": "0.4.6",
-        "@lancedb/vectordb-win32-x64-msvc": "0.4.6"
+        "@lancedb/vectordb-darwin-arm64": "0.21.2-beta.0",
+        "@lancedb/vectordb-darwin-x64": "0.21.2-beta.0",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.21.2-beta.0",
+        "@lancedb/vectordb-linux-x64-gnu": "0.21.2-beta.0",
+        "@lancedb/vectordb-win32-x64-msvc": "0.21.2-beta.0"
+      },
+      "peerDependencies": {
+        "@apache-arrow/ts": "^14.0.2",
+        "apache-arrow": "^14.0.2"
       }
     },
     "../node/node_modules/apache-arrow": {

docs/requirements.txt

@@ -1,6 +1,7 @@
 mkdocs==1.5.3
 mkdocs-jupyter==0.24.1
 mkdocs-material==9.5.3
-mkdocstrings[python]==0.20.0
+mkdocstrings[python]==0.25.2
+griffe
 mkdocs-render-swagger-plugin
 pydantic

docs/src/ann_indexes.md

@@ -18,25 +18,24 @@ See the [indexing](concepts/index_ivfpq.md) concepts guide for more information
 Lance supports `IVF_PQ` index type by default.
 
 === "Python"
+    === "Sync API"
 
-    Creating indexes is done via the [create_index](https://lancedb.github.io/lancedb/python/#lancedb.table.LanceTable.create_index) method.
+        Creating indexes is done via the [create_index](https://lancedb.github.io/lancedb/python/#lancedb.table.LanceTable.create_index) method.
 
-    ```python
-    import lancedb
-    import numpy as np
-    uri = "data/sample-lancedb"
-    db = lancedb.connect(uri)
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb"
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-numpy"
+        --8<-- "python/python/tests/docs/test_guide_index.py:create_ann_index"
+        ```
+    === "Async API"
+        Creating indexes is done via the [create_index](https://lancedb.github.io/lancedb/python/#lancedb.table.LanceTable.create_index) method.
 
-    # Create 10,000 sample vectors
-    data = [{"vector": row, "item": f"item {i}"}
-        for i, row in enumerate(np.random.random((10_000, 1536)).astype('float32'))]
-
-    # Add the vectors to a table
-    tbl = db.create_table("my_vectors", data=data)
-
-    # Create and train the index - you need to have enough data in the table for an effective training step
-    tbl.create_index(num_partitions=256, num_sub_vectors=96)
-    ```
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb"
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-numpy"
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb-ivfpq"
+        --8<-- "python/python/tests/docs/test_guide_index.py:create_ann_index_async"
+        ```
 
 === "TypeScript"
 
@@ -45,9 +44,9 @@ Lance supports `IVF_PQ` index type by default.
         Creating indexes is done via the [lancedb.Table.createIndex](../js/classes/Table.md/#createIndex) method.
 
         ```typescript
-        --8<--- "nodejs/examples/ann_indexes.ts:import"
+        --8<--- "nodejs/examples/ann_indexes.test.ts:import"
 
-        --8<-- "nodejs/examples/ann_indexes.ts:ingest"
+        --8<-- "nodejs/examples/ann_indexes.test.ts:ingest"
         ```
 
     === "vectordb (deprecated)"
@@ -70,7 +69,7 @@ Lance supports `IVF_PQ` index type by default.
 
 The following IVF_PQ paramters can be specified:
 
-- **distance_type**: The distance metric to use. By default it uses euclidean distance "`L2`".
+- **distance_type**: The distance metric to use. By default it uses euclidean distance "`l2`".
   We also support "cosine" and "dot" distance as well.
 - **num_partitions**: The number of partitions in the index. The default is the square root
   of the number of rows.
@@ -83,6 +82,7 @@ The following IVF_PQ paramters can be specified:
 - **num_sub_vectors**: The number of sub-vectors (M) that will be created during Product Quantization (PQ).
   For D dimensional vector, it will be divided into `M` subvectors with dimension `D/M`, each of which is replaced by
   a single PQ code. The default is the dimension of the vector divided by 16.
+- **num_bits**: The number of bits used to encode each sub-vector. Only 4 and 8 are supported. The higher the number of bits, the higher the accuracy of the index, also the slower search. The default is 8.
 
 !!! note
 
@@ -126,7 +126,9 @@ You can specify the GPU device to train IVF partitions via
         accelerator="mps"
     )
     ```
-
+!!! note
+    GPU based indexing is not yet supported with our asynchronous client.
+    
 Troubleshooting:
 
 If you see `AssertionError: Torch not compiled with CUDA enabled`, you need to [install
@@ -140,23 +142,27 @@ There are a couple of parameters that can be used to fine-tune the search:
 
 - **limit** (default: 10): The amount of results that will be returned
 - **nprobes** (default: 20): The number of probes used. A higher number makes search more accurate but also slower.<br/>
-  Most of the time, setting nprobes to cover 5-10% of the dataset should achieve high recall with low latency.<br/>
-  e.g., for 1M vectors divided up into 256 partitions, nprobes should be set to ~20-40.<br/>
-  Note: nprobes is only applicable if an ANN index is present. If specified on a table without an ANN index, it is ignored.
+  Most of the time, setting nprobes to cover 5-15% of the dataset should achieve high recall with low latency.<br/>
+    - _For example_, For a dataset of 1 million vectors divided into 256 partitions, `nprobes` should be set to ~20-40. This value can be adjusted to achieve the optimal balance between search latency and search quality. <br/>
+
 - **refine_factor** (default: None): Refine the results by reading extra elements and re-ranking them in memory.<br/>
   A higher number makes search more accurate but also slower. If you find the recall is less than ideal, try refine_factor=10 to start.<br/>
-  e.g., for 1M vectors divided into 256 partitions, if you're looking for top 20, then refine_factor=200 reranks the whole partition.<br/>
-  Note: refine_factor is only applicable if an ANN index is present. If specified on a table without an ANN index, it is ignored.
+    - _For example_, For a dataset of 1 million vectors divided into 256 partitions, setting the `refine_factor` to 200 will initially retrieve the top 4,000 candidates (top k * refine_factor) from all searched partitions. These candidates are then reranked to determine the final top 20 results.<br/>
+!!! note
+    Both `nprobes` and `refine_factor` are only applicable if an ANN index is present. If specified on a table without an ANN index, those parameters are ignored.
+
 
 === "Python"
+    === "Sync API"
 
-    ```python
-    tbl.search(np.random.random((1536))) \
-        .limit(2) \
-        .nprobes(20) \
-        .refine_factor(10) \
-        .to_pandas()
-    ```
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_async"
+        ```
 
     ```text
                                               vector       item       _distance
@@ -169,7 +175,7 @@ There are a couple of parameters that can be used to fine-tune the search:
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/ann_indexes.ts:search1"
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search1"
         ```
 
     === "vectordb (deprecated)"
@@ -193,17 +199,23 @@ The search will return the data requested in addition to the distance of each it
 You can further filter the elements returned by a search using a where clause.
 
 === "Python"
+    === "Sync API"
 
-    ```python
-    tbl.search(np.random.random((1536))).where("item != 'item 1141'").to_pandas()
-    ```
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_with_filter"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_async_with_filter"
+        ```
 
 === "TypeScript"
 
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/ann_indexes.ts:search2"
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search2"
         ```
 
     === "vectordb (deprecated)"
@@ -218,10 +230,16 @@ You can select the columns returned by the query using a select clause.
 
 === "Python"
 
-    ```python
-    tbl.search(np.random.random((1536))).select(["vector"]).to_pandas()
-    ```
+    === "Sync API"
 
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_with_select"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_async_with_select"
+        ```
 
     ```text
                                                 vector _distance
@@ -235,7 +253,7 @@ You can select the columns returned by the query using a select clause.
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/ann_indexes.ts:search3"
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search3"
         ```
 
     === "vectordb (deprecated)"
@@ -273,9 +291,17 @@ Product quantization can lead to approximately `16 * sizeof(float32) / 1 = 64` t
 
 `num_partitions` is used to decide how many partitions the first level `IVF` index uses.
 Higher number of partitions could lead to more efficient I/O during queries and better accuracy, but it takes much more time to train.
-On `SIFT-1M` dataset, our benchmark shows that keeping each partition 1K-4K rows lead to a good latency / recall.
+On `SIFT-1M` dataset, our benchmark shows that keeping each partition 4K-8K rows lead to a good latency / recall.
 
-`num_sub_vectors` specifies how many Product Quantization (PQ) short codes to generate on each vector. Because
+`num_sub_vectors` specifies how many Product Quantization (PQ) short codes to generate on each vector. The number should be a factor of the vector dimension. Because
 PQ is a lossy compression of the original vector, a higher `num_sub_vectors` usually results in
-less space distortion, and thus yields better accuracy. However, a higher `num_sub_vectors` also causes heavier I/O and
-more PQ computation, and thus, higher latency. `dimension / num_sub_vectors` should be a multiple of 8 for optimum SIMD efficiency.
+less space distortion, and thus yields better accuracy. However, a higher `num_sub_vectors` also causes heavier I/O and more PQ computation, and thus, higher latency. `dimension / num_sub_vectors` should be a multiple of 8 for optimum SIMD efficiency.
+
+!!! note
+    if `num_sub_vectors` is set to be greater than the vector dimension, you will see errors like `attempt to divide by zero`
+
+### How to choose `m` and `ef_construction` for `IVF_HNSW_*` index?
+
+`m` determines the number of connections a new node establishes with its closest neighbors upon entering the graph. Typically, `m` falls within the range of 5 to 48. Lower `m` values are suitable for low-dimensional data or scenarios where recall is less critical. Conversely, higher `m` values are beneficial for high-dimensional data or when high recall is required. In essence, a larger `m` results in a denser graph with increased connectivity, but at the expense of higher memory consumption.
+
+`ef_construction` balances build speed and accuracy. Higher values increase accuracy but slow down the build process. A typical range is 150 to 300. For good search results, a minimum value of 100 is recommended. In most cases, setting this value above 500 offers no additional benefit. Ensure that `ef_construction` is always set to a value equal to or greater than `ef` in the search phase

docs/src/ann_indexes.ts

@@ -3,6 +3,7 @@ import * as vectordb from "vectordb";
 // --8<-- [end:import]
 
 (async () => {
+  console.log("ann_indexes.ts: start");
   // --8<-- [start:ingest]
   const db = await vectordb.connect("data/sample-lancedb");
 
@@ -49,5 +50,5 @@ import * as vectordb from "vectordb";
     .execute();
   // --8<-- [end:search3]
 
-  console.log("Ann indexes: done");
+  console.log("ann_indexes.ts: done");
 })();

docs/src/api_reference.md

@@ -4,5 +4,5 @@ The API reference for the LanceDB client SDKs are available at the following loc
 
 - [Python](python/python.md)
 - [JavaScript (legacy vectordb package)](javascript/modules.md)
-- [JavaScript (newer @lancedb/lancedb package)](js/modules.md)
+- [JavaScript (newer @lancedb/lancedb package)](js/globals.md)
 - [Rust](https://docs.rs/lancedb/latest/lancedb/index.html)

docs/src/assets/colab.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="117" height="20"><linearGradient id="b" x2="0" y2="100%"><stop offset="0" stop-color="#bbb" stop-opacity=".1"/><stop offset="1" stop-opacity=".1"/></linearGradient><clipPath id="a"><rect width="117" height="20" rx="3" fill="#fff"/></clipPath><g clip-path="url(#a)"><path fill="#555" d="M0 0h30v20H0z"/><path fill="#007ec6" d="M30 0h87v20H30z"/><path fill="url(#b)" d="M0 0h117v20H0z"/></g><g fill="#fff" text-anchor="middle" font-family="DejaVu Sans,Verdana,Geneva,sans-serif" font-size="110"><svg x="4px" y="0px" width="22px" height="20px" viewBox="-2 0 28 24" style="background-color: #fff;border-radius: 1px;"><path style="fill:#e8710a;" d="M1.977,16.77c-2.667-2.277-2.605-7.079,0-9.357C2.919,8.057,3.522,9.075,4.49,9.691c-1.152,1.6-1.146,3.201-0.004,4.803C3.522,15.111,2.918,16.126,1.977,16.77z"/><path style="fill:#f9ab00;" d="M12.257,17.114c-1.767-1.633-2.485-3.658-2.118-6.02c0.451-2.91,2.139-4.893,4.946-5.678c2.565-0.718,4.964-0.217,6.878,1.819c-0.884,0.743-1.707,1.547-2.434,2.446C18.488,8.827,17.319,8.435,16,8.856c-2.404,0.767-3.046,3.241-1.494,5.644c-0.241,0.275-0.493,0.541-0.721,0.826C13.295,15.939,12.511,16.3,12.257,17.114z"/><path style="fill:#e8710a;" d="M19.529,9.682c0.727-0.899,1.55-1.703,2.434-2.446c2.703,2.783,2.701,7.031-0.005,9.764c-2.648,2.674-6.936,2.725-9.701,0.115c0.254-0.814,1.038-1.175,1.528-1.788c0.228-0.285,0.48-0.552,0.721-0.826c1.053,0.916,2.254,1.268,3.6,0.83C20.502,14.551,21.151,11.927,19.529,9.682z"/><path style="fill:#f9ab00;" d="M4.49,9.691C3.522,9.075,2.919,8.057,1.977,7.413c2.209-2.398,5.721-2.942,8.476-1.355c0.555,0.32,0.719,0.606,0.285,1.128c-0.157,0.188-0.258,0.422-0.391,0.631c-0.299,0.47-0.509,1.067-0.929,1.371C8.933,9.539,8.523,8.847,8.021,8.746C6.673,8.475,5.509,8.787,4.49,9.691z"/><path style="fill:#f9ab00;" d="M1.977,16.77c0.941-0.644,1.545-1.659,2.509-2.277c1.373,1.152,2.85,1.433,4.45,0.499c0.332-0.194,0.503-0.088,0.673,0.19c0.386,0.635,0.753,1.285,1.181,1.89c0.34,0.48,0.222,0.715-0.253,1.006C7.84,19.73,4.205,19.188,1.977,16.77z"/></svg><text x="245" y="140" transform="scale(.1)" textLength="30"> </text><text x="725" y="150" fill="#010101" fill-opacity=".3" transform="scale(.1)" textLength="770">Open in Colab</text><text x="725" y="140" transform="scale(.1)" textLength="770">Open in Colab</text></g> </svg>

docs/src/assets/ghost.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="88.25" height="28" role="img" aria-label="GHOST"><title>GHOST</title><g shape-rendering="crispEdges"><rect width="88.25" height="28" fill="#000"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="100"><image x="9" y="7" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text transform="scale(.1)" x="541.25" y="175" textLength="442.5" fill="#fff" font-weight="bold">GHOST</text></g></svg>

docs/src/assets/github.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="95.5" height="28" role="img" aria-label="GITHUB"><title>GITHUB</title><g shape-rendering="crispEdges"><rect width="95.5" height="28" fill="#121011"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="100"><image x="9" y="7" width="14" height="14" xlink:href="data:image/svg+xml;base64,PHN2ZyBmaWxsPSJ3aGl0ZSIgcm9sZT0iaW1nIiB2aWV3Qm94PSIwIDAgMjQgMjQiIHhtbG5zPSJodHRwOi8vd3d3LnczLm9yZy8yMDAwL3N2ZyI+PHRpdGxlPkdpdEh1YjwvdGl0bGU+PHBhdGggZD0iTTEyIC4yOTdjLTYuNjMgMC0xMiA1LjM3My0xMiAxMiAwIDUuMzAzIDMuNDM4IDkuOCA4LjIwNSAxMS4zODUuNi4xMTMuODItLjI1OC44Mi0uNTc3IDAtLjI4NS0uMDEtMS4wNC0uMDE1LTIuMDQtMy4zMzguNzI0LTQuMDQyLTEuNjEtNC4wNDItMS42MUM0LjQyMiAxOC4wNyAzLjYzMyAxNy43IDMuNjMzIDE3LjdjLTEuMDg3LS43NDQuMDg0LS43MjkuMDg0LS43MjkgMS4yMDUuMDg0IDEuODM4IDEuMjM2IDEuODM4IDEuMjM2IDEuMDcgMS44MzUgMi44MDkgMS4zMDUgMy40OTUuOTk4LjEwOC0uNzc2LjQxNy0xLjMwNS43Ni0xLjYwNS0yLjY2NS0uMy01LjQ2Ni0xLjMzMi01LjQ2Ni01LjkzIDAtMS4zMS40NjUtMi4zOCAxLjIzNS0zLjIyLS4xMzUtLjMwMy0uNTQtMS41MjMuMTA1LTMuMTc2IDAgMCAxLjAwNS0uMzIyIDMuMyAxLjIzLjk2LS4yNjcgMS45OC0uMzk5IDMtLjQwNSAxLjAyLjAwNiAyLjA0LjEzOCAzIC40MDUgMi4yOC0xLjU1MiAzLjI4NS0xLjIzIDMuMjg1LTEuMjMuNjQ1IDEuNjUzLjI0IDIuODczLjEyIDMuMTc2Ljc2NS44NCAxLjIzIDEuOTEgMS4yMyAzLjIyIDAgNC42MS0yLjgwNSA1LjYyNS01LjQ3NSA1LjkyLjQyLjM2LjgxIDEuMDk2LjgxIDIuMjIgMCAxLjYwNi0uMDE1IDIuODk2LS4wMTUgMy4yODYgMCAuMzE1LjIxLjY5LjgyNS41N0MyMC41NjUgMjIuMDkyIDI0IDE3LjU5MiAyNCAxMi4yOTdjMC02LjYyNy01LjM3My0xMi0xMi0xMiIvPjwvc3ZnPg=="/><text transform="scale(.1)" x="577.5" y="175" textLength="515" fill="#fff" font-weight="bold">GITHUB</text></g></svg>

docs/src/assets/open_hf_space.svg

@@ -0,0 +1,22 @@
+<svg width="147" height="20" viewBox="0 0 147 20" fill="none" xmlns="http://www.w3.org/2000/svg">
+<rect x="0.5" y="0.5" width="145.482" height="19" rx="9.5" fill="white" stroke="#EFEFEF"/>
+<path d="M14.1863 10.9251V12.7593H16.0205V10.9251H14.1863Z" fill="#FF3270"/>
+<path d="M17.8707 10.9251V12.7593H19.7049V10.9251H17.8707Z" fill="#861FFF"/>
+<path d="M14.1863 7.24078V9.07496H16.0205V7.24078H14.1863Z" fill="#097EFF"/>
+<path fill-rule="evenodd" clip-rule="evenodd" d="M12.903 6.77179C12.903 6.32194 13.2676 5.95728 13.7175 5.95728C14.1703 5.95728 15.2556 5.95728 16.1094 5.95728C16.7538 5.95728 17.2758 6.47963 17.2758 7.12398V9.6698H19.8217C20.4661 9.6698 20.9884 10.1922 20.9884 10.8365C20.9884 11.6337 20.9884 12.4309 20.9884 13.2282C20.9884 13.678 20.6237 14.0427 20.1738 14.0427H17.3039H16.5874H13.7175C13.2676 14.0427 12.903 13.678 12.903 13.2282V9.71653V9.64174V6.77179ZM14.1863 7.24066V9.07485H16.0205V7.24066H14.1863ZM14.1863 12.7593V10.9251H16.0205V12.7593H14.1863ZM17.8708 12.7593V10.9251H19.705V12.7593H17.8708Z" fill="black"/>
+<path d="M18.614 8.35468L20.7796 6.18905M20.7796 6.18905V7.66073M20.7796 6.18905L19.2724 6.18905" stroke="black" stroke-width="0.686298" stroke-linecap="round" stroke-linejoin="round"/>
+<path d="M31.6082 13.9838C30.8546 13.9838 30.1895 13.802 29.6132 13.4385C29.0368 13.066 28.5846 12.5429 28.2565 11.869C27.9373 11.1862 27.7777 10.3749 27.7777 9.43501C27.7777 8.49511 27.9373 7.69265 28.2565 7.02762C28.5846 6.3626 29.0368 5.85275 29.6132 5.49807C30.1895 5.14339 30.8546 4.96605 31.6082 4.96605C32.3708 4.96605 33.0403 5.14339 33.6166 5.49807C34.193 5.85275 34.6408 6.3626 34.96 7.02762C35.2881 7.69265 35.4521 8.49511 35.4521 9.43501C35.4521 10.3749 35.2881 11.1862 34.96 11.869C34.6408 12.5429 34.193 13.066 33.6166 13.4385C33.0403 13.802 32.3708 13.9838 31.6082 13.9838ZM31.6082 12.6404C32.291 12.6404 32.8363 12.3523 33.2442 11.7759C33.6521 11.1907 33.856 10.4104 33.856 9.43501C33.856 8.45964 33.6521 7.69708 33.2442 7.14733C32.8363 6.58871 32.291 6.3094 31.6082 6.3094C30.9255 6.3094 30.3802 6.58871 29.9723 7.14733C29.5644 7.69708 29.3605 8.45964 29.3605 9.43501C29.3605 10.4104 29.5644 11.1907 29.9723 11.7759C30.3802 12.3523 30.9255 12.6404 31.6082 12.6404Z" fill="#2C3236"/>
+<path d="M37.0592 16.4045V7.29363H38.3227L38.4291 7.98526H38.4823C38.7572 7.75472 39.0631 7.55521 39.4 7.38674C39.7459 7.21826 40.0961 7.13403 40.4508 7.13403C41.2665 7.13403 41.8961 7.43551 42.3395 8.03846C42.7917 8.64142 43.0178 9.44831 43.0178 10.4591C43.0178 11.204 42.8848 11.8424 42.6188 12.3744C42.3528 12.8976 42.0069 13.2966 41.5813 13.5715C41.1646 13.8463 40.7124 13.9838 40.2247 13.9838C39.9409 13.9838 39.6572 13.9217 39.3734 13.7976C39.0897 13.6646 38.8148 13.4872 38.5488 13.2656L38.5887 14.3562V16.4045H37.0592ZM39.9055 12.7202C40.3399 12.7202 40.7035 12.5296 40.9961 12.1483C41.2887 11.767 41.435 11.2084 41.435 10.4724C41.435 9.81629 41.3242 9.30644 41.1025 8.94289C40.8808 8.57935 40.5217 8.39757 40.0252 8.39757C39.5641 8.39757 39.0853 8.64142 38.5887 9.1291V12.1749C38.8281 12.37 39.0587 12.5119 39.2803 12.6005C39.502 12.6803 39.7104 12.7202 39.9055 12.7202Z" fill="#2C3236"/>
+<path d="M47.3598 13.9838C46.7568 13.9838 46.2115 13.8508 45.7238 13.5848C45.2361 13.3099 44.8504 12.9197 44.5667 12.4143C44.2829 11.9 44.141 11.2838 44.141 10.5656C44.141 9.85619 44.2829 9.24437 44.5667 8.73009C44.8593 8.2158 45.2361 7.82122 45.6972 7.54634C46.1583 7.27147 46.6415 7.13403 47.147 7.13403C47.741 7.13403 48.2376 7.26703 48.6366 7.53304C49.0356 7.79018 49.3371 8.15373 49.541 8.62368C49.745 9.08476 49.847 9.62122 49.847 10.233C49.847 10.5523 49.8248 10.8005 49.7805 10.9779H45.6307C45.7016 11.5542 45.91 12.002 46.2558 12.3212C46.6016 12.6404 47.0361 12.8 47.5593 12.8C47.843 12.8 48.1046 12.7601 48.344 12.6803C48.5923 12.5917 48.8361 12.472 49.0755 12.3212L49.5942 13.2789C49.2839 13.4828 48.9381 13.6513 48.5568 13.7843C48.1755 13.9173 47.7765 13.9838 47.3598 13.9838ZM45.6174 9.94043H48.5169C48.5169 9.43501 48.4061 9.04043 48.1844 8.75669C47.9627 8.46408 47.6302 8.31777 47.1869 8.31777C46.8056 8.31777 46.4642 8.45964 46.1627 8.74339C45.8701 9.01826 45.6883 9.41728 45.6174 9.94043Z" fill="#2C3236"/>
+<path d="M51.3078 13.8242V7.29363H52.5714L52.6778 8.17147H52.731C53.0236 7.88772 53.3428 7.64388 53.6886 7.43994C54.0344 7.236 54.429 7.13403 54.8724 7.13403C55.5728 7.13403 56.0827 7.36014 56.4019 7.81235C56.7211 8.26457 56.8807 8.90299 56.8807 9.72762V13.8242H55.3512V9.92713C55.3512 9.38624 55.2714 9.00496 55.1118 8.78329C54.9522 8.56161 54.6906 8.45078 54.327 8.45078C54.0433 8.45078 53.7906 8.52171 53.5689 8.66358C53.3561 8.79659 53.1123 8.99609 52.8374 9.2621V13.8242H51.3078Z" fill="#2C3236"/>
+<path d="M61.4131 13.8242V7.29363H62.9426V13.8242H61.4131ZM62.1845 6.14979C61.9096 6.14979 61.6879 6.06999 61.5195 5.91038C61.351 5.75078 61.2668 5.53797 61.2668 5.27196C61.2668 5.01482 61.351 4.80644 61.5195 4.64684C61.6879 4.48723 61.9096 4.40743 62.1845 4.40743C62.4594 4.40743 62.6811 4.48723 62.8495 4.64684C63.018 4.80644 63.1022 5.01482 63.1022 5.27196C63.1022 5.53797 63.018 5.75078 62.8495 5.91038C62.6811 6.06999 62.4594 6.14979 62.1845 6.14979Z" fill="#2C3236"/>
+<path d="M64.8941 13.8242V7.29363H66.1576L66.264 8.17147H66.3172C66.6098 7.88772 66.929 7.64388 67.2748 7.43994C67.6207 7.236 68.0152 7.13403 68.4586 7.13403C69.1591 7.13403 69.6689 7.36014 69.9881 7.81235C70.3074 8.26457 70.467 8.90299 70.467 9.72762V13.8242H68.9374V9.92713C68.9374 9.38624 68.8576 9.00496 68.698 8.78329C68.5384 8.56161 68.2768 8.45078 67.9133 8.45078C67.6295 8.45078 67.3768 8.52171 67.1551 8.66358C66.9423 8.79659 66.6985 8.99609 66.4236 9.2621V13.8242H64.8941Z" fill="#2C3236"/>
+<path d="M75.1323 13.8242V5.12565H76.6752V8.62368H80.1998V5.12565H81.7427V13.8242H80.1998V9.96703H76.6752V13.8242H75.1323Z" fill="#2C3236"/>
+<path d="M83.9517 13.8242V5.12565H89.2054V6.4291H85.4945V8.88969H88.6601V10.1931H85.4945V13.8242H83.9517Z" fill="#2C3236"/>
+<path d="M95.9349 13.9838C95.3497 13.9838 94.7822 13.8729 94.2324 13.6513C93.6915 13.4296 93.2127 13.1148 92.796 12.7069L93.7004 11.6562C94.0108 11.9488 94.3654 12.1882 94.7645 12.3744C95.1635 12.5518 95.5625 12.6404 95.9615 12.6404C96.458 12.6404 96.8349 12.5385 97.092 12.3345C97.3492 12.1306 97.4778 11.8601 97.4778 11.5232C97.4778 11.1596 97.3492 10.8981 97.092 10.7385C96.8438 10.5789 96.5245 10.4148 96.1344 10.2463L94.9374 9.72762C94.6536 9.60348 94.3743 9.44388 94.0994 9.2488C93.8334 9.05373 93.6117 8.80546 93.4344 8.50398C93.2659 8.2025 93.1817 7.83895 93.1817 7.41334C93.1817 6.95225 93.3058 6.53994 93.5541 6.17639C93.8113 5.80398 94.1571 5.51137 94.5915 5.29856C95.0349 5.07689 95.5403 4.96605 96.1078 4.96605C96.6132 4.96605 97.1009 5.06802 97.5709 5.27196C98.0408 5.46703 98.4442 5.73304 98.7812 6.06999L97.9965 7.05423C97.7216 6.82368 97.429 6.64191 97.1186 6.5089C96.8172 6.3759 96.4802 6.3094 96.1078 6.3094C95.6999 6.3094 95.3674 6.4025 95.1103 6.58871C94.862 6.76605 94.7379 7.01432 94.7379 7.33353C94.7379 7.55521 94.7999 7.74142 94.9241 7.89215C95.0571 8.03403 95.23 8.15816 95.4428 8.26457C95.6556 8.36211 95.8817 8.45964 96.1211 8.55718L97.3048 9.0493C97.8191 9.27097 98.2403 9.56358 98.5684 9.92713C98.8965 10.2818 99.0605 10.7739 99.0605 11.4035C99.0605 11.8734 98.9364 12.3035 98.6881 12.6936C98.4398 13.0838 98.0807 13.3986 97.6108 13.638C97.1497 13.8685 96.591 13.9838 95.9349 13.9838Z" fill="#2C3236"/>
+<path d="M100.509 16.4045V7.29363H101.773L101.879 7.98526H101.932C102.207 7.75472 102.513 7.55521 102.85 7.38674C103.196 7.21826 103.546 7.13403 103.901 7.13403C104.717 7.13403 105.346 7.43551 105.79 8.03846C106.242 8.64142 106.468 9.44831 106.468 10.4591C106.468 11.204 106.335 11.8424 106.069 12.3744C105.803 12.8976 105.457 13.2966 105.031 13.5715C104.615 13.8463 104.162 13.9838 103.675 13.9838C103.391 13.9838 103.107 13.9217 102.824 13.7976C102.54 13.6646 102.265 13.4872 101.999 13.2656L102.039 14.3562V16.4045H100.509ZM103.356 12.7202C103.79 12.7202 104.154 12.5296 104.446 12.1483C104.739 11.767 104.885 11.2084 104.885 10.4724C104.885 9.81629 104.774 9.30644 104.553 8.94289C104.331 8.57935 103.972 8.39757 103.475 8.39757C103.014 8.39757 102.535 8.64142 102.039 9.1291V12.1749C102.278 12.37 102.509 12.5119 102.73 12.6005C102.952 12.6803 103.16 12.7202 103.356 12.7202Z" fill="#2C3236"/>
+<path d="M109.444 13.9838C108.876 13.9838 108.411 13.8064 108.047 13.4518C107.692 13.0971 107.515 12.636 107.515 12.0685C107.515 11.368 107.821 10.8271 108.433 10.4458C109.045 10.0557 110.02 9.78969 111.359 9.64782C111.35 9.30201 111.257 9.00496 111.08 8.75669C110.911 8.49954 110.605 8.37097 110.162 8.37097C109.843 8.37097 109.528 8.43304 109.218 8.55718C108.916 8.68132 108.619 8.83206 108.326 9.0094L107.768 7.98526C108.131 7.75472 108.539 7.55521 108.991 7.38674C109.452 7.21826 109.94 7.13403 110.454 7.13403C111.27 7.13403 111.878 7.37787 112.277 7.86555C112.685 8.34437 112.888 9.04043 112.888 9.95373V13.8242H111.625L111.518 13.1059H111.465C111.173 13.3542 110.858 13.5626 110.521 13.7311C110.193 13.8995 109.834 13.9838 109.444 13.9838ZM109.936 12.7867C110.202 12.7867 110.441 12.7247 110.654 12.6005C110.876 12.4675 111.111 12.2902 111.359 12.0685V10.6055C110.472 10.7207 109.856 10.8936 109.51 11.1242C109.164 11.3458 108.991 11.6207 108.991 11.9488C108.991 12.2414 109.08 12.4542 109.257 12.5872C109.435 12.7202 109.661 12.7867 109.936 12.7867Z" fill="#2C3236"/>
+<path d="M117.446 13.9838C116.851 13.9838 116.315 13.8508 115.836 13.5848C115.366 13.3099 114.989 12.9197 114.706 12.4143C114.431 11.9 114.293 11.2838 114.293 10.5656C114.293 9.83846 114.444 9.2222 114.746 8.71679C115.047 8.2025 115.446 7.81235 115.943 7.54634C116.448 7.27147 116.989 7.13403 117.565 7.13403C117.982 7.13403 118.346 7.20496 118.656 7.34684C118.966 7.48871 119.241 7.66161 119.48 7.86555L118.736 8.86309C118.567 8.71235 118.394 8.59708 118.217 8.51728C118.04 8.42861 117.849 8.38427 117.645 8.38427C117.122 8.38427 116.692 8.58378 116.355 8.98279C116.027 9.38181 115.863 9.9094 115.863 10.5656C115.863 11.2128 116.022 11.736 116.342 12.135C116.67 12.534 117.091 12.7335 117.605 12.7335C117.862 12.7335 118.102 12.6803 118.323 12.5739C118.554 12.4587 118.762 12.3256 118.948 12.1749L119.574 13.1857C119.272 13.4518 118.935 13.6513 118.563 13.7843C118.19 13.9173 117.818 13.9838 117.446 13.9838Z" fill="#2C3236"/>
+<path d="M123.331 13.9838C122.728 13.9838 122.183 13.8508 121.695 13.5848C121.207 13.3099 120.822 12.9197 120.538 12.4143C120.254 11.9 120.112 11.2838 120.112 10.5656C120.112 9.85619 120.254 9.24437 120.538 8.73009C120.83 8.2158 121.207 7.82122 121.668 7.54634C122.13 7.27147 122.613 7.13403 123.118 7.13403C123.712 7.13403 124.209 7.26703 124.608 7.53304C125.007 7.79018 125.308 8.15373 125.512 8.62368C125.716 9.08476 125.818 9.62122 125.818 10.233C125.818 10.5523 125.796 10.8005 125.752 10.9779H121.602C121.673 11.5542 121.881 12.002 122.227 12.3212C122.573 12.6404 123.007 12.8 123.53 12.8C123.814 12.8 124.076 12.7601 124.315 12.6803C124.563 12.5917 124.807 12.472 125.047 12.3212L125.565 13.2789C125.255 13.4828 124.909 13.6513 124.528 13.7843C124.147 13.9173 123.748 13.9838 123.331 13.9838ZM121.589 9.94043H124.488C124.488 9.43501 124.377 9.04043 124.156 8.75669C123.934 8.46408 123.601 8.31777 123.158 8.31777C122.777 8.31777 122.435 8.45964 122.134 8.74339C121.841 9.01826 121.66 9.41728 121.589 9.94043Z" fill="#2C3236"/>
+<path d="M129.101 13.9838C128.658 13.9838 128.215 13.8995 127.771 13.7311C127.328 13.5537 126.947 13.3365 126.627 13.0793L127.346 12.0951C127.638 12.3168 127.931 12.4941 128.223 12.6271C128.516 12.7601 128.826 12.8266 129.154 12.8266C129.509 12.8266 129.771 12.7513 129.939 12.6005C130.108 12.4498 130.192 12.2636 130.192 12.0419C130.192 11.8557 130.121 11.705 129.979 11.5897C129.846 11.4656 129.673 11.3591 129.46 11.2705C129.248 11.1729 129.026 11.0798 128.795 10.9912C128.512 10.8848 128.228 10.7562 127.944 10.6055C127.669 10.4458 127.443 10.2463 127.266 10.0069C127.088 9.75866 127 9.45274 127 9.0892C127 8.51284 127.213 8.04289 127.638 7.67935C128.064 7.3158 128.64 7.13403 129.367 7.13403C129.828 7.13403 130.241 7.21383 130.604 7.37344C130.968 7.53304 131.282 7.71482 131.548 7.91876L130.844 8.84979C130.613 8.68132 130.378 8.54831 130.139 8.45078C129.908 8.34437 129.664 8.29117 129.407 8.29117C129.079 8.29117 128.835 8.36211 128.676 8.50398C128.516 8.63698 128.436 8.80545 128.436 9.0094C128.436 9.26654 128.569 9.46161 128.835 9.59462C129.101 9.72762 129.412 9.85619 129.766 9.98033C130.068 10.0867 130.36 10.2197 130.644 10.3793C130.928 10.5301 131.163 10.7296 131.349 10.9779C131.544 11.2261 131.642 11.5542 131.642 11.9621C131.642 12.5207 131.424 12.9995 130.99 13.3986C130.555 13.7887 129.926 13.9838 129.101 13.9838Z" fill="#2C3236"/>
+</svg>

docs/src/assets/python.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" width="97.5" height="28" role="img" aria-label="PYTHON"><title>PYTHON</title><g shape-rendering="crispEdges"><rect width="97.5" height="28" fill="#3670a0"/></g><g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="100"><image x="9" y="7" width="14" height="14" xlink:href="data:image/svg+xml;base64,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"/><text transform="scale(.1)" x="587.5" y="175" textLength="535" fill="#fff" font-weight="bold">PYTHON</text></g></svg>

docs/src/basic.md

@@ -35,6 +35,15 @@
             }
             })
             ```
+
+        !!! note "Yarn users"
+
+            Unlike other package managers, Yarn does not automatically resolve peer dependencies. If you are using Yarn, you will need to manually install 'apache-arrow':
+
+            ```shell
+            yarn add apache-arrow
+            ```
+
     === "vectordb (deprecated)"
 
         ```shell
@@ -53,6 +62,15 @@
             }
             })
             ```
+
+        !!! note "Yarn users"
+
+            Unlike other package managers, Yarn does not automatically resolve peer dependencies. If you are using Yarn, you will need to manually install 'apache-arrow':
+
+            ```shell
+            yarn add apache-arrow
+            ```
+
 === "Rust"
 
     ```shell
@@ -115,21 +133,22 @@ recommend switching to stable releases.
 ## Connect to a database
 
 === "Python"
+    === "Sync API"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:imports"
-    --8<-- "python/python/tests/docs/test_basic.py:connect"
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:imports"
 
-    --8<-- "python/python/tests/docs/test_basic.py:connect_async"
-    ```
+        --8<-- "python/python/tests/docs/test_basic.py:set_uri"
+        --8<-- "python/python/tests/docs/test_basic.py:connect"
+        ```
+    === "Async API"
 
-    !!! note "Asynchronous Python API"
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:imports"
 
-        The asynchronous Python API is new and has some slight differences compared
-        to the synchronous API.  Feel free to start using the asynchronous version.
-        Once all features have migrated we will start to move the synchronous API to
-        use the same syntax as the asynchronous API.  To help with this migration we
-        have created a [migration guide](migration.md) detailing the differences.
+        --8<-- "python/python/tests/docs/test_basic.py:set_uri"
+        --8<-- "python/python/tests/docs/test_basic.py:connect_async"
+        ```
 
 === "Typescript[^1]"
 
@@ -139,7 +158,7 @@ recommend switching to stable releases.
         import * as lancedb from "@lancedb/lancedb";
         import * as arrow from "apache-arrow";
 
-        --8<-- "nodejs/examples/basic.ts:connect"
+        --8<-- "nodejs/examples/basic.test.ts:connect"
         ```
 
     === "vectordb (deprecated)"
@@ -173,28 +192,40 @@ table.
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:create_table"
-    --8<-- "python/python/tests/docs/test_basic.py:create_table_async"
-    ```
-
     If the table already exists, LanceDB will raise an error by default.
     If you want to overwrite the table, you can pass in `mode="overwrite"`
     to the `create_table` method.
 
-    You can also pass in a pandas DataFrame directly:
+    === "Sync API"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:create_table_pandas"
-    --8<-- "python/python/tests/docs/test_basic.py:create_table_async_pandas"
-    ```
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_table"
+        ```
+
+        You can also pass in a pandas DataFrame directly:
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_table_pandas"
+        ```
+
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_table_async"
+        ```
+
+        You can also pass in a pandas DataFrame directly:
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_table_async_pandas"
+        ```
 
 === "Typescript[^1]"
 
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:create_table"
+        --8<-- "nodejs/examples/basic.test.ts:create_table"
         ```
 
     === "vectordb (deprecated)"
@@ -237,10 +268,16 @@ similar to a `CREATE TABLE` statement in SQL.
 
 === "Python"
 
-      ```python
-      --8<-- "python/python/tests/docs/test_basic.py:create_empty_table"
-      --8<-- "python/python/tests/docs/test_basic.py:create_empty_table_async"
-      ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_empty_table"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_empty_table_async"
+        ```
 
     !!! note "You can define schema in Pydantic"
         LanceDB comes with Pydantic support, which allows you to define the schema of your data using Pydantic models. This makes it easy to work with LanceDB tables and data. Learn more about all supported types in [tables guide](./guides/tables.md).
@@ -250,7 +287,7 @@ similar to a `CREATE TABLE` statement in SQL.
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:create_empty_table"
+        --8<-- "nodejs/examples/basic.test.ts:create_empty_table"
         ```
 
     === "vectordb (deprecated)"
@@ -271,16 +308,22 @@ Once created, you can open a table as follows:
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:open_table"
-    --8<-- "python/python/tests/docs/test_basic.py:open_table_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:open_table"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:open_table_async"
+        ```
 
 === "Typescript[^1]"
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:open_table"
+        --8<-- "nodejs/examples/basic.test.ts:open_table"
         ```
 
     === "vectordb (deprecated)"
@@ -300,16 +343,22 @@ If you forget the name of your table, you can always get a listing of all table
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:table_names"
-    --8<-- "python/python/tests/docs/test_basic.py:table_names_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:table_names"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:table_names_async"
+        ```
 
 === "Typescript[^1]"
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:table_names"
+        --8<-- "nodejs/examples/basic.test.ts:table_names"
         ```
 
     === "vectordb (deprecated)"
@@ -330,16 +379,22 @@ After a table has been created, you can always add more data to it as follows:
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:add_data"
-    --8<-- "python/python/tests/docs/test_basic.py:add_data_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:add_data"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:add_data_async"
+        ```
 
 === "Typescript[^1]"
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:add_data"
+        --8<-- "nodejs/examples/basic.test.ts:add_data"
         ```
 
     === "vectordb (deprecated)"
@@ -360,10 +415,16 @@ Once you've embedded the query, you can find its nearest neighbors as follows:
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:vector_search"
-    --8<-- "python/python/tests/docs/test_basic.py:vector_search_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:vector_search"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:vector_search_async"
+        ```
 
     This returns a pandas DataFrame with the results.
 
@@ -371,7 +432,7 @@ Once you've embedded the query, you can find its nearest neighbors as follows:
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:vector_search"
+        --8<-- "nodejs/examples/basic.test.ts:vector_search"
         ```
 
     === "vectordb (deprecated)"
@@ -402,16 +463,22 @@ LanceDB allows you to create an ANN index on a table as follows:
 
 === "Python"
 
-    ```py
-    --8<-- "python/python/tests/docs/test_basic.py:create_index"
-    --8<-- "python/python/tests/docs/test_basic.py:create_index_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_index"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:create_index_async"
+        ```
 
 === "Typescript[^1]"
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:create_index"
+        --8<-- "nodejs/examples/basic.test.ts:create_index"
         ```
 
     === "vectordb (deprecated)"
@@ -441,17 +508,23 @@ This can delete any number of rows that match the filter.
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:delete_rows"
-    --8<-- "python/python/tests/docs/test_basic.py:delete_rows_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:delete_rows"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:delete_rows_async"
+        ```
 
 === "Typescript[^1]"
 
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:delete_rows"
+        --8<-- "nodejs/examples/basic.test.ts:delete_rows"
         ```
 
     === "vectordb (deprecated)"
@@ -473,7 +546,10 @@ simple or complex as needed. To see what expressions are supported, see the
 
 === "Python"
 
-      Read more: [lancedb.table.Table.delete][]
+    === "Sync API"
+        Read more: [lancedb.table.Table.delete][]
+    === "Async API"
+        Read more: [lancedb.table.AsyncTable.delete][]
 
 === "Typescript[^1]"
 
@@ -495,10 +571,16 @@ Use the `drop_table()` method on the database to remove a table.
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_basic.py:drop_table"
-    --8<-- "python/python/tests/docs/test_basic.py:drop_table_async"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:drop_table"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:drop_table_async"
+        ```
 
     This permanently removes the table and is not recoverable, unlike deleting rows.
     By default, if the table does not exist an exception is raised. To suppress this,
@@ -509,7 +591,7 @@ Use the `drop_table()` method on the database to remove a table.
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/basic.ts:drop_table"
+        --8<-- "nodejs/examples/basic.test.ts:drop_table"
         ```
 
     === "vectordb (deprecated)"
@@ -533,18 +615,25 @@ You can use the embedding API when working with embedding models. It automatical
 
 === "Python"
 
-    ```python
-    --8<-- "python/python/tests/docs/test_embeddings_optional.py:imports"
-    --8<-- "python/python/tests/docs/test_embeddings_optional.py:openai_embeddings"
-    ```
+    === "Sync API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_embeddings_optional.py:imports"
+
+        --8<-- "python/python/tests/docs/test_embeddings_optional.py:openai_embeddings"
+        ```
+    === "Async API"
+
+        Coming soon to the async API.
+        https://github.com/lancedb/lancedb/issues/1938
 
 === "Typescript[^1]"
 
     === "@lancedb/lancedb"
 
         ```typescript
-        --8<-- "nodejs/examples/embedding.ts:imports"
-        --8<-- "nodejs/examples/embedding.ts:openai_embeddings"
+        --8<-- "nodejs/examples/embedding.test.ts:imports"
+        --8<-- "nodejs/examples/embedding.test.ts:openai_embeddings"
         ```
 
 === "Rust"
@@ -554,7 +643,7 @@ You can use the embedding API when working with embedding models. It automatical
     --8<-- "rust/lancedb/examples/openai.rs:openai_embeddings"
     ```
 
-Learn about using the existing integrations and creating custom embedding functions in the [embedding API guide](./embeddings/).
+Learn about using the existing integrations and creating custom embedding functions in the [embedding API guide](./embeddings/index.md).
 
 
 ## What's next

docs/src/basic_legacy.ts

@@ -1,6 +1,14 @@
 // --8<-- [start:import]
 import * as lancedb from "vectordb";
-import { Schema, Field, Float32, FixedSizeList, Int32, Float16 } from "apache-arrow";
+import {
+  Schema,
+  Field,
+  Float32,
+  FixedSizeList,
+  Int32,
+  Float16,
+} from "apache-arrow";
+import * as arrow from "apache-arrow";
 // --8<-- [end:import]
 import * as fs from "fs";
 import { Table as ArrowTable, Utf8 } from "apache-arrow";
@@ -20,10 +28,33 @@ const example = async () => {
       { vector: [3.1, 4.1], item: "foo", price: 10.0 },
       { vector: [5.9, 26.5], item: "bar", price: 20.0 },
     ],
-    { writeMode: lancedb.WriteMode.Overwrite }
+    { writeMode: lancedb.WriteMode.Overwrite },
   );
   // --8<-- [end:create_table]
-
+  {
+    // --8<-- [start:create_table_with_schema]
+    const schema = new arrow.Schema([
+      new arrow.Field(
+        "vector",
+        new arrow.FixedSizeList(
+          2,
+          new arrow.Field("item", new arrow.Float32(), true),
+        ),
+      ),
+      new arrow.Field("item", new arrow.Utf8(), true),
+      new arrow.Field("price", new arrow.Float32(), true),
+    ]);
+    const data = [
+      { vector: [3.1, 4.1], item: "foo", price: 10.0 },
+      { vector: [5.9, 26.5], item: "bar", price: 20.0 },
+    ];
+    const tbl = await db.createTable({
+      name: "myTableWithSchema",
+      data,
+      schema,
+    });
+    // --8<-- [end:create_table_with_schema]
+  }
 
   // --8<-- [start:add]
   const newData = Array.from({ length: 500 }, (_, i) => ({
@@ -43,38 +74,39 @@ const example = async () => {
   // --8<-- [end:create_index]
 
   // --8<-- [start:create_empty_table]
-  const schema = new Schema([
-    new Field("id", new Int32()),
-    new Field("name", new Utf8()),
+  const schema = new arrow.Schema([
+    new arrow.Field("id", new arrow.Int32()),
+    new arrow.Field("name", new arrow.Utf8()),
   ]);
+
   const empty_tbl = await db.createTable({ name: "empty_table", schema });
   // --8<-- [end:create_empty_table]
-
-  // --8<-- [start:create_f16_table]
-  const dim = 16
-  const total = 10
-  const f16_schema = new Schema([
-      new Field('id', new Int32()),
+  {
+    // --8<-- [start:create_f16_table]
+    const dim = 16;
+    const total = 10;
+    const schema = new Schema([
+      new Field("id", new Int32()),
       new Field(
-        'vector',
-        new FixedSizeList(dim, new Field('item', new Float16(), true)),
-        false
-      )
-    ])
-  const data = lancedb.makeArrowTable(
+        "vector",
+        new FixedSizeList(dim, new Field("item", new Float16(), true)),
+        false,
+      ),
+    ]);
+    const data = lancedb.makeArrowTable(
       Array.from(Array(total), (_, i) => ({
         id: i,
-        vector: Array.from(Array(dim), Math.random)
+        vector: Array.from(Array(dim), Math.random),
       })),
-      { f16_schema }
-    )
-  const table = await db.createTable('f16_tbl', data)
-  // --8<-- [end:create_f16_table]
+      { schema },
+    );
+    const table = await db.createTable("f16_tbl", data);
+    // --8<-- [end:create_f16_table]
+  }
 
   // --8<-- [start:search]
   const query = await tbl.search([100, 100]).limit(2).execute();
   // --8<-- [end:search]
-  console.log(query);
 
   // --8<-- [start:delete]
   await tbl.delete('item = "fizz"');
@@ -86,8 +118,9 @@ const example = async () => {
 };
 
 async function main() {
+  console.log("basic_legacy.ts: start");
   await example();
-  console.log("Basic example: done");
+  console.log("basic_legacy.ts: done");
 }
 
 main();

docs/src/cloud/cloud_faq.md

@@ -0,0 +1,34 @@
+This section provides answers to the most common questions asked about LanceDB Cloud. By following these guidelines, you can ensure a smooth, performant experience with LanceDB Cloud.
+
+### Should I reuse the database connection?
+Yes! It is recommended to establish a single database connection and maintain it throughout your interaction with the tables within. 
+
+LanceDB uses HTTP connections to communicate with the servers. By re-using the Connection object, you avoid the overhead of repeatedly establishing HTTP connections, significantly improving efficiency.
+
+### Should I re-use the `Table` object?
+`table = db.open_table()` should be called once and used for all subsequent table operations. If there are changes to the opened table, `table` always reflect the **latest version** of the data. 
+
+### What should I do if I need to search for rows by `id`?
+LanceDB Cloud currently does not support an ID or primary key column. You are recommended to add a 
+user-defined ID column. To significantly improve the query performance with SQL causes, a scalar BITMAP/BTREE index should be created on this column. 
+
+### What are the vector indexing types supported by LanceDB Cloud?
+We support `IVF_PQ` and `IVF_HNSW_SQ` as the `index_type` which is passed to `create_index`. LanceDB Cloud tunes the indexing parameters automatically to achieve the best tradeoff between query latency and query quality.
+
+### When I add new rows to a table, do I need to manually update the index?
+No! LanceDB Cloud triggers an asynchronous background job to index the new vectors.
+
+Even though indexing is asynchronous, your vectors will still be immediately searchable. LanceDB uses brute-force search to search over unindexed rows. This makes you new data is immediately available, but does increase latency temporarily. To disable the brute-force part of search, set the `fast_search` flag in your query to `true`.
+
+### Do I need to reindex the whole dataset if only a small portion of the data is deleted or updated?
+No! Similar to adding data to the table, LanceDB Cloud triggers an asynchronous background job to update the existing indices. Therefore, no action is needed from users and there is absolutely no 
+downtime expected.
+
+### How do I know whether an index has been created?
+While index creation in LanceDB Cloud is generally fast, querying immediately after a `create_index` call may result in errors. It's recommended to use `list_indices` to verify index creation before querying.
+
+### Why is my query latency higher than expected?
+Multiple factors can impact query latency. To reduce query latency, consider the following:
+- Send pre-warm queries: send a few queries to warm up the cache before an actual user query.
+- Check network latency: LanceDB Cloud is hosted in AWS `us-east-1` region. It is recommended to run queries from an EC2 instance that is in the same region.
+- Create scalar indices: If you are filtering on metadata, it is recommended to create scalar indices on those columns. This will speedup searches with metadata filtering. See [here](../guides/scalar_index.md) for more details on creating a scalar index.

docs/src/cloud/index.md

@@ -2,7 +2,7 @@
 
 LanceDB Cloud is a SaaS (software-as-a-service) solution that runs serverless in the cloud, clearly separating storage from compute. It's designed to be highly scalable without breaking the bank. LanceDB Cloud is currently in private beta with general availability coming soon, but you can apply for early access with the private beta release by signing up below.
 
-[Try out LanceDB Cloud](https://noteforms.com/forms/lancedb-mailing-list-cloud-kty1o5?notionforms=1&utm_source=notionforms){ .md-button .md-button--primary }
+[Try out LanceDB Cloud (Public Beta)](https://cloud.lancedb.com){ .md-button .md-button--primary }
 
 ## Architecture
 

docs/src/concepts/data_management.md

@@ -13,7 +13,7 @@ The following concepts are important to keep in mind:
 - Data is versioned, with each insert operation creating a new version of the dataset and an update to the manifest that tracks versions via metadata
 
 !!! note
-    1. First, each version contains metadata and just the new/updated data in your transaction. So if you have 100 versions, they aren't 100 duplicates of the same data. However, they do have 100x the metadata overhead of a single version, which can result in slower queries.  
+    1. First, each version contains metadata and just the new/updated data in your transaction. So if you have 100 versions, they aren't 100 duplicates of the same data. However, they do have 100x the metadata overhead of a single version, which can result in slower queries.
     2. Second, these versions exist to keep LanceDB scalable and consistent. We do not immediately blow away old versions when creating new ones because other clients might be in the middle of querying the old version. It's important to retain older versions for as long as they might be queried.
 
 ## What are fragments?
@@ -37,6 +37,10 @@ Depending on the use case and dataset, optimal compaction will have different re
 - It’s always better to use *batch* inserts rather than adding 1 row at a time (to avoid too small fragments). If single-row inserts are unavoidable, run compaction on a regular basis to merge them into larger fragments.
 - Keep the number of fragments under 100, which is suitable for most use cases (for *really* large datasets of >500M rows, more fragments might be needed)
 
+!!! note
+
+    LanceDB Cloud/Enterprise supports [auto-compaction](https://docs.lancedb.com/enterprise/architecture/architecture#write-path) which automatically optimizes fragments in the background as data changes.
+
 ## Deletion
 
 Although Lance allows you to delete rows from a dataset, it does not actually delete the data immediately. It simply marks the row as deleted in the `DataFile` that represents a fragment. For a given version of the dataset, each fragment can have up to one deletion file (if no rows were ever deleted from that fragment, it will not have a deletion file). This is important to keep in mind because it means that the data is still there, and can be recovered if needed, as long as that version still exists based on your backup policy.
@@ -50,13 +54,9 @@ Reindexing is the process of updating the index to account for new data, keeping
 
 Both LanceDB OSS and Cloud support reindexing, but the process (at least for now) is different for each, depending on the type of index.
 
-When a reindex job is triggered in the background, the entire data is reindexed, but in the interim as new queries come in, LanceDB will combine results from the existing index with exhaustive kNN search on the new data. This is done to ensure that you're still searching on all your data, but it does come at a performance cost. The more data that you add without reindexing, the impact on latency (due to exhaustive search) can be noticeable.
+In LanceDB OSS, re-indexing happens synchronously when you call either `create_index` or `optimize` on a table. In LanceDB Cloud, re-indexing happens asynchronously as you add and update data in your table.
 
-### Vector reindex
+By default, queries will search new data even if it has yet to be indexed. This is done using brute-force methods, such as kNN for vector search, and combined with the fast index search results. This is done to ensure that you're always searching over all your data, but it does come at a performance cost. Without reindexing, adding more data to a table will make queries slower and more expensive. This behavior can be disabled by setting the [fast_search](https://lancedb.github.io/lancedb/python/python/#lancedb.query.AsyncQuery.fast_search) parameter which will instruct the query to ignore un-indexed data.
 
-* LanceDB Cloud supports incremental reindexing, where a background process will trigger a new index build for you automatically when new data is added to a dataset
+* LanceDB Cloud/Enterprise supports [automatic incremental reindexing](https://docs.lancedb.com/core#vector-index) for vector, scalar, and FTS indices, where a background process will trigger a new index build for you automatically when new data is added or modified in a dataset
 * LanceDB OSS requires you to manually trigger a reindex operation -- we are working on adding incremental reindexing to LanceDB OSS as well
-
-### FTS reindex
-
-FTS reindexing is supported in both LanceDB OSS and Cloud, but requires that it's manually rebuilt once you have a significant enough amount of new data added that needs to be reindexed. We [updated](https://github.com/lancedb/lancedb/pull/762) Tantivy's default heap size from 128MB to 1GB in LanceDB to make it much faster to reindex, by up to 10x from the default settings.

docs/src/concepts/index_hnsw.md

@@ -0,0 +1,99 @@
+
+# Understanding HNSW index
+
+Approximate Nearest Neighbor (ANN) search is a method for finding data points near a given point in a dataset, though not always the exact nearest one. HNSW is one of the most accurate and fastest Approximate Nearest Neighbour search algorithms, It’s beneficial in high-dimensional spaces where finding the same nearest neighbor would be too slow and costly
+
+[Jump to usage](#usage)
+There are three main types of ANN search algorithms:
+
+* **Tree-based search algorithms**: Use a tree structure to organize and store data points.
+* **Hash-based search algorithms**: Use a specialized geometric hash table to store and manage data points. These algorithms typically focus on theoretical guarantees, and don't usually perform as well as the other approaches in practice.
+* **Graph-based search algorithms**: Use a graph structure to store data points, which can be a bit complex. 
+
+HNSW is a graph-based algorithm. All graph-based search algorithms rely on the idea of a k-nearest neighbor (or k-approximate nearest neighbor) graph, which we outline below.  
+HNSW also combines this with the ideas behind a classic 1-dimensional search data structure: the skip list.
+
+## k-Nearest Neighbor Graphs and k-approximate Nearest neighbor Graphs
+The k-nearest neighbor graph actually predates its use for ANN search. Its construction is quite simple:
+
+* Each vector in the dataset is given an associated vertex.
+* Each vertex has outgoing edges to its k nearest neighbors. That is, the k closest other vertices by Euclidean distance between the two corresponding vectors. This can be thought of as a "friend list" for the vertex.
+* For some applications (including nearest-neighbor search), the incoming edges are also added.
+
+Eventually, it was realized that the following greedy search method over such a graph typically results in good approximate nearest neighbors:
+
+* Given a query vector, start at some fixed "entry point" vertex (e.g. the approximate center node).
+* Look at that vertex's neighbors. If any of them are closer to the query vector than the current vertex, then move to that vertex.
+* Repeat until a local optimum is found.
+
+The above algorithm also generalizes to e.g. top 10 approximate nearest neighbors.
+
+Computing a k-nearest neighbor graph is actually quite slow, taking quadratic time in the dataset size. It was quickly realized that near-identical performance can be achieved using a k-approximate nearest neighbor graph. That is, instead of obtaining the k-nearest neighbors for each vertex, an approximate nearest neighbor search data structure is used to build much faster.  
+In fact, another data structure is not needed: This can be done "incrementally".
+That is, if you start with a k-ANN graph for n-1 vertices, you can extend it to a k-ANN graph for n vertices as well by using the graph to obtain the k-ANN for the new vertex.
+
+One downside of k-NN and k-ANN graphs alone is that one must typically build them with a large value of k to get decent results, resulting in a large index.
+
+
+## HNSW: Hierarchical Navigable Small Worlds
+
+HNSW builds on k-ANN in two main ways:
+
+* Instead of getting the k-approximate nearest neighbors for a large value of k, it sparsifies the k-ANN graph using a carefully chosen "edge pruning" heuristic, allowing for the number of edges per vertex to be limited to a relatively small constant.
+* The "entry point" vertex is chosen dynamically using a recursively constructed data structure on a subset of the data, similarly to a skip list.
+
+This recursive structure can be thought of as separating into layers:
+
+* At the bottom-most layer, an k-ANN graph on the whole dataset is present.
+* At the second layer, a k-ANN graph on a fraction of the dataset (e.g. 10%) is present.
+* At the Lth layer, a k-ANN graph is present. It is over a (constant) fraction (e.g. 10%) of the vectors/vertices present in the L-1th layer.
+
+Then the greedy search routine operates as follows:
+
+* At the top layer (using an arbitrary vertex as an entry point), use the greedy local search routine on the k-ANN graph to get an approximate nearest neighbor at that layer.
+* Using the approximate nearest neighbor found in the previous layer as an entry point, find an approximate nearest neighbor in the next layer with the same method.
+* Repeat until the bottom-most layer is reached. Then use the entry point to find multiple nearest neighbors (e.g. top 10).
+
+
+## Usage
+
+There are three key parameters to set when constructing an HNSW index:
+
+* `metric`: Use an `l2` euclidean distance metric. We also support `dot` and `cosine` distance.
+* `m`: The number of neighbors to select for each vector in the HNSW graph.
+* `ef_construction`: The number of candidates to evaluate during the construction of the HNSW graph.
+
+
+We can combine the above concepts to understand how to build and query an HNSW index in LanceDB.
+
+### Construct index
+
+```python
+import lancedb
+import numpy as np
+uri = "/tmp/lancedb"
+db = lancedb.connect(uri)
+
+# Create 10,000 sample vectors
+data = [
+    {"vector": row, "item": f"item {i}"}
+    for i, row in enumerate(np.random.random((10_000, 1536)).astype('float32'))
+]
+
+# Add the vectors to a table
+tbl = db.create_table("my_vectors", data=data)
+
+# Create and train the HNSW index for a 1536-dimensional vector
+# Make sure you have enough data in the table for an effective training step
+tbl.create_index(index_type=IVF_HNSW_SQ)
+
+```
+
+### Query the index
+
+```python
+# Search using a random 1536-dimensional embedding
+tbl.search(np.random.random((1536))) \
+    .limit(2) \
+    .to_pandas()
+```

docs/src/concepts/index_ivfpq.md

@@ -47,7 +47,7 @@ We can combine the above concepts to understand how to build and query an IVF-PQ
 
 There are three key parameters to set when constructing an IVF-PQ index:
 
-* `metric`: Use an `L2` euclidean distance metric. We also support `dot` and `cosine` distance.
+* `metric`: Use an `l2` euclidean distance metric. We also support `dot` and `cosine` distance.
 * `num_partitions`: The number of partitions in the IVF portion of the index.
 * `num_sub_vectors`: The number of sub-vectors that will be created during Product Quantization (PQ).
 
@@ -56,10 +56,12 @@ In Python, the index can be created as follows:
 ```python
 # Create and train the index for a 1536-dimensional vector
 # Make sure you have enough data in the table for an effective training step
-tbl.create_index(metric="L2", num_partitions=256, num_sub_vectors=96)
+tbl.create_index(metric="l2", num_partitions=256, num_sub_vectors=96)
 ```
+!!! note
+    `num_partitions`=256 and `num_sub_vectors`=96 does not work for every dataset. Those values needs to be adjusted for your particular dataset.
 
-The `num_partitions` is usually chosen to target a particular number of vectors per partition. `num_sub_vectors` is typically chosen based on the desired recall and the dimensionality of the vector. See the [FAQs](#faq) below for best practices on choosing these parameters.
+The `num_partitions` is usually chosen to target a particular number of vectors per partition. `num_sub_vectors` is typically chosen based on the desired recall and the dimensionality of the vector. See [here](../ann_indexes.md/#how-to-choose-num_partitions-and-num_sub_vectors-for-ivf_pq-index) for best practices on choosing these parameters.
 
 
 ### Query the index

docs/src/embeddings/available_embedding_models/multimodal_embedding_functions/imagebind_embedding.md

@@ -0,0 +1,67 @@
+# Imagebind embeddings
+We have support for [imagebind](https://github.com/facebookresearch/ImageBind) model embeddings. You can download our version of the packaged model via - `pip install imagebind-packaged==0.1.2`.
+
+This function is registered as `imagebind` and supports Audio, Video and Text modalities(extending to Thermal,Depth,IMU data):
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `"imagebind_huge"` | Name of the model. |
+| `device` | `str` | `"cpu"` | The device to run the model on. Can be `"cpu"` or `"gpu"`. |
+| `normalize` | `bool` | `False` | set to `True` to normalize your inputs before model ingestion. |
+
+Below is an example demonstrating how the API works:
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+db = lancedb.connect(tmp_path)
+func = get_registry().get("imagebind").create()
+
+class ImageBindModel(LanceModel):
+    text: str
+    image_uri: str = func.SourceField()
+    audio_path: str
+    vector: Vector(func.ndims()) = func.VectorField()
+
+# add locally accessible image paths
+text_list=["A dog.", "A car", "A bird"]
+image_paths=[".assets/dog_image.jpg", ".assets/car_image.jpg", ".assets/bird_image.jpg"]
+audio_paths=[".assets/dog_audio.wav", ".assets/car_audio.wav", ".assets/bird_audio.wav"]
+
+# Load data
+inputs = [
+    {"text": a, "audio_path": b, "image_uri": c}
+    for a, b, c in zip(text_list, audio_paths, image_paths)
+]
+
+#create table and add data
+table = db.create_table("img_bind", schema=ImageBindModel)
+table.add(inputs)
+```
+
+Now, we can search using any modality:
+
+#### image search
+```python
+query_image = "./assets/dog_image2.jpg" #download an image and enter that path here
+actual = table.search(query_image).limit(1).to_pydantic(ImageBindModel)[0]
+print(actual.text == "dog")
+```
+#### audio search
+
+```python
+query_audio = "./assets/car_audio2.wav" #download an audio clip and enter path here
+actual = table.search(query_audio).limit(1).to_pydantic(ImageBindModel)[0]
+print(actual.text == "car")
+```
+#### Text search
+You can add any input query and fetch the result as follows:
+```python
+query = "an animal which flies and tweets" 
+actual = table.search(query).limit(1).to_pydantic(ImageBindModel)[0]
+print(actual.text == "bird")
+```
+
+If you have any questions about the embeddings API, supported models, or see a relevant model missing, please raise an issue [on GitHub](https://github.com/lancedb/lancedb/issues).

docs/src/embeddings/available_embedding_models/multimodal_embedding_functions/jina_multimodal_embedding.md

@@ -0,0 +1,51 @@
+# Jina Embeddings : Multimodal
+
+Jina embeddings can also be used to embed both text and image data, only some of the models support image data and you can check the list
+under [https://jina.ai/embeddings/](https://jina.ai/embeddings/)
+
+Supported parameters (to be passed in `create` method) are:
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `"jina-clip-v1"` | The model ID of the jina model to use |
+
+Usage Example:
+
+```python
+    import os
+    import requests
+    import lancedb
+    from lancedb.pydantic import LanceModel, Vector
+    from lancedb.embeddings import get_registry
+    import pandas as pd
+
+    os.environ['JINA_API_KEY'] = 'jina_*'
+
+    db = lancedb.connect("~/.lancedb")
+    func = get_registry().get("jina").create()
+
+
+    class Images(LanceModel):
+        label: str
+        image_uri: str = func.SourceField()  # image uri as the source
+        image_bytes: bytes = func.SourceField()  # image bytes as the source
+        vector: Vector(func.ndims()) = func.VectorField()  # vector column
+        vec_from_bytes: Vector(func.ndims()) = func.VectorField()  # Another vector column
+
+
+    table = db.create_table("images", schema=Images)
+    labels = ["cat", "cat", "dog", "dog", "horse", "horse"]
+    uris = [
+        "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
+        "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
+        "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
+        "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
+        "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
+        "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
+    ]
+    # get each uri as bytes
+    image_bytes = [requests.get(uri).content for uri in uris]
+    table.add(
+      pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": image_bytes})
+    )
+```

docs/src/embeddings/available_embedding_models/multimodal_embedding_functions/openclip_embedding.md

@@ -0,0 +1,82 @@
+# OpenClip embeddings
+We support CLIP model embeddings using the open source alternative, [open-clip](https://github.com/mlfoundations/open_clip) which supports various customizations. It is registered as `open-clip` and supports the following customizations:
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `"ViT-B-32"` | The name of the model. |
+| `pretrained` | `str` | `"laion2b_s34b_b79k"` | The name of the pretrained model to load. |
+| `device` | `str` | `"cpu"` | The device to run the model on. Can be `"cpu"` or `"gpu"`. |
+| `batch_size` | `int` | `64` | The number of images to process in a batch. |
+| `normalize` | `bool` | `True` | Whether to normalize the input images before feeding them to the model. |
+
+This embedding function supports ingesting images as both bytes and urls. You can query them using both test and other images.
+
+!!! info
+    LanceDB supports ingesting images directly from accessible links.
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+db = lancedb.connect(tmp_path)
+func = get_registry().get("open-clip").create()
+
+class Images(LanceModel):
+    label: str
+    image_uri: str = func.SourceField() # image uri as the source
+    image_bytes: bytes = func.SourceField() # image bytes as the source
+    vector: Vector(func.ndims()) = func.VectorField() # vector column 
+    vec_from_bytes: Vector(func.ndims()) = func.VectorField() # Another vector column 
+
+table = db.create_table("images", schema=Images)
+labels = ["cat", "cat", "dog", "dog", "horse", "horse"]
+uris = [
+    "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
+    "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
+    "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
+    "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
+    "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
+    "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
+]
+# get each uri as bytes
+image_bytes = [requests.get(uri).content for uri in uris]
+table.add(
+    pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": image_bytes})
+)
+```
+Now we can search using text from both the default vector column and the custom vector column
+```python
+
+# text search
+actual = table.search("man's best friend").limit(1).to_pydantic(Images)[0]
+print(actual.label) # prints "dog"
+
+frombytes = (
+    table.search("man's best friend", vector_column_name="vec_from_bytes")
+    .limit(1)
+    .to_pydantic(Images)[0]
+)
+print(frombytes.label)
+
+```
+
+Because we're using a multi-modal embedding function, we can also search using images
+
+```python
+# image search
+query_image_uri = "http://farm1.staticflickr.com/200/467715466_ed4a31801f_z.jpg"
+image_bytes = requests.get(query_image_uri).content
+query_image = Image.open(io.BytesIO(image_bytes))
+actual = table.search(query_image).limit(1).to_pydantic(Images)[0]
+print(actual.label == "dog")
+
+# image search using a custom vector column
+other = (
+    table.search(query_image, vector_column_name="vec_from_bytes")
+    .limit(1)
+    .to_pydantic(Images)[0]
+)
+print(actual.label)
+
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/aws_bedrock_embedding.md

@@ -0,0 +1,51 @@
+# AWS Bedrock Text Embedding Functions
+
+AWS Bedrock supports multiple base models for generating text embeddings. You need to setup the AWS credentials to use this embedding function.
+You can do so by using `awscli` and also add your session_token:
+```shell
+aws configure
+aws configure set aws_session_token "<your_session_token>"
+```
+to ensure that the credentials are set up correctly, you can run the following command:
+```shell
+aws sts get-caller-identity
+```
+
+Supported Embedding modelIDs are:
+* `amazon.titan-embed-text-v1`
+* `cohere.embed-english-v3`
+* `cohere.embed-multilingual-v3`
+
+Supported parameters (to be passed in `create` method) are:
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| **name** | str | "amazon.titan-embed-text-v1" | The model ID of the bedrock model to use. Supported base models for Text Embeddings: amazon.titan-embed-text-v1, cohere.embed-english-v3, cohere.embed-multilingual-v3 |
+| **region** | str | "us-east-1" | Optional name of the AWS Region in which the service should be called (e.g., "us-east-1"). |
+| **profile_name** | str | None | Optional name of the AWS profile to use for calling the Bedrock service. If not specified, the default profile will be used. |
+| **assumed_role** | str | None | Optional ARN of an AWS IAM role to assume for calling the Bedrock service. If not specified, the current active credentials will be used. |
+| **role_session_name** | str | "lancedb-embeddings" | Optional name of the AWS IAM role session to use for calling the Bedrock service. If not specified, a "lancedb-embeddings" name will be used. |
+| **runtime** | bool | True | Optional choice of getting different client to perform operations with the Amazon Bedrock service. |
+| **max_retries** | int | 7 | Optional number of retries to perform when a request fails. |
+
+Usage Example:
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+import pandas as pd
+
+model = get_registry().get("bedrock-text").create()
+
+class TextModel(LanceModel):
+    text: str = model.SourceField()
+    vector: Vector(model.ndims()) = model.VectorField()
+
+df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
+db = lancedb.connect("tmp_path")
+tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+tbl.add(df)
+rs = tbl.search("hello").limit(1).to_pandas()
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/cohere_embedding.md

@@ -0,0 +1,63 @@
+# Cohere Embeddings
+
+Using cohere API requires cohere package, which can be installed using `pip install cohere`. Cohere embeddings are used to generate embeddings for text data. The embeddings can be used for various tasks like semantic search, clustering, and classification.
+You also need to set the `COHERE_API_KEY` environment variable to use the Cohere API.
+
+Supported models are:
+
+- embed-english-v3.0
+- embed-multilingual-v3.0
+- embed-english-light-v3.0
+- embed-multilingual-light-v3.0
+- embed-english-v2.0
+- embed-english-light-v2.0
+- embed-multilingual-v2.0
+
+
+Supported parameters (to be passed in `create` method) are:
+
+| Parameter | Type | Default Value | Description |
+|---|---|--------|---------|
+| `name` | `str` | `"embed-english-v2.0"` | The model ID of the cohere model to use. Supported base models for Text Embeddings: embed-english-v3.0, embed-multilingual-v3.0, embed-english-light-v3.0, embed-multilingual-light-v3.0, embed-english-v2.0, embed-english-light-v2.0, embed-multilingual-v2.0 |
+| `source_input_type` | `str` | `"search_document"` | The type of input data to be used for the source column. |
+| `query_input_type` | `str` | `"search_query"` | The type of input data to be used for the query. |
+
+Cohere supports following input types:
+
+| Input Type               | Description                          |
+|-------------------------|---------------------------------------|
+| "`search_document`"     | Used for embeddings stored in a vector|
+|                         | database for search use-cases.        |
+| "`search_query`"        | Used for embeddings of search queries |
+|                         | run against a vector DB               |
+| "`semantic_similarity`" | Specifies the given text will be used |
+|                         | for Semantic Textual Similarity (STS) |
+| "`classification`"      | Used for embeddings passed through a  |
+|                         | text classifier.                      |
+| "`clustering`"          | Used for the embeddings run through a |
+|                         | clustering algorithm                  |
+
+Usage Example:
+    
+```python
+    import lancedb
+    from lancedb.pydantic import LanceModel, Vector
+    from lancedb.embeddings import EmbeddingFunctionRegistry
+
+    cohere = EmbeddingFunctionRegistry
+        .get_instance()
+        .get("cohere")
+        .create(name="embed-multilingual-v2.0")
+
+    class TextModel(LanceModel):
+        text: str = cohere.SourceField()
+        vector: Vector(cohere.ndims()) =  cohere.VectorField()
+
+    data = [ { "text": "hello world" },
+            { "text": "goodbye world" }]
+
+    db = lancedb.connect("~/.lancedb")
+    tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+    tbl.add(data)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/gemini_embedding.md

@@ -0,0 +1,35 @@
+# Gemini Embeddings
+With Google's Gemini, you can represent text (words, sentences, and blocks of text) in a vectorized form, making it easier to compare and contrast embeddings. For example, two texts that share a similar subject matter or sentiment should have similar embeddings, which can be identified through mathematical comparison techniques such as cosine similarity. For more on how and why you should use embeddings, refer to the Embeddings guide.
+The Gemini Embedding Model API supports various task types:
+
+| Task Type               | Description                                                                                                                                                |
+|-------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| "`retrieval_query`"     | Specifies the given text is a query in a search/retrieval setting.                                                                                         |
+| "`retrieval_document`"  | Specifies the given text is a document in a search/retrieval setting. Using this task type requires a title but is automatically proided by Embeddings API |
+| "`semantic_similarity`" | Specifies the given text will be used for Semantic Textual Similarity (STS).                                                                               |
+| "`classification`"      | Specifies that the embeddings will be used for classification.                                                                                             |
+| "`clusering`"           | Specifies that the embeddings will be used for clustering.                                                                                                 |
+
+
+Usage Example:
+
+```python
+import lancedb
+import pandas as pd
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+
+model = get_registry().get("gemini-text").create()
+
+class TextModel(LanceModel):
+    text: str = model.SourceField()
+    vector: Vector(model.ndims()) = model.VectorField()
+
+df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
+db = lancedb.connect("~/.lancedb")
+tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+tbl.add(df)
+rs = tbl.search("hello").limit(1).to_pandas()
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/huggingface_embedding.md

@@ -0,0 +1,24 @@
+# Huggingface embedding models
+We offer support for all Hugging Face models (which can be loaded via [transformers](https://huggingface.co/docs/transformers/en/index) library). The default model is `colbert-ir/colbertv2.0` which also has its own special callout - `registry.get("colbert")`. Some Hugging Face models might require custom models defined on the HuggingFace Hub in their own modeling files. You may enable this by setting `trust_remote_code=True`. This option should only be set to True for repositories you trust and in which you have read the code, as it will execute code present on the Hub on your local machine. 
+
+Example usage - 
+```python
+import lancedb
+import pandas as pd
+
+from lancedb.embeddings import get_registry
+from lancedb.pydantic import LanceModel, Vector
+
+model = get_registry().get("huggingface").create(name='facebook/bart-base')
+
+class Words(LanceModel):
+    text: str = model.SourceField()
+    vector: Vector(model.ndims()) = model.VectorField()
+
+df = pd.DataFrame({"text": ["hi hello sayonara", "goodbye world"]})
+table = db.create_table("greets", schema=Words)
+table.add(df)
+query = "old greeting"
+actual = table.search(query).limit(1).to_pydantic(Words)[0]
+print(actual.text)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/ibm_watsonx_ai_embedding.md

@@ -0,0 +1,75 @@
+# IBM watsonx.ai Embeddings
+
+Generate text embeddings using IBM's watsonx.ai platform.
+
+## Supported Models
+
+You can find a list of supported models at [IBM watsonx.ai Documentation](https://dataplatform.cloud.ibm.com/docs/content/wsj/analyze-data/fm-models-embed.html?context=wx). The currently supported model names are:
+
+- `ibm/slate-125m-english-rtrvr`
+- `ibm/slate-30m-english-rtrvr`
+- `sentence-transformers/all-minilm-l12-v2`
+- `intfloat/multilingual-e5-large`
+
+## Parameters
+
+The following parameters can be passed to the `create` method:
+
+| Parameter  | Type     | Default Value                    | Description                                               |
+|------------|----------|----------------------------------|-----------------------------------------------------------|
+| name       | str      | "ibm/slate-125m-english-rtrvr"   | The model ID of the watsonx.ai model to use               |
+| api_key    | str      | None                             | Optional IBM Cloud API key (or set `WATSONX_API_KEY`)     |
+| project_id | str      | None                             | Optional watsonx project ID (or set `WATSONX_PROJECT_ID`) |
+| url        | str      | None                             | Optional custom URL for the watsonx.ai instance           |
+| params     | dict     | None                             | Optional additional parameters for the embedding model    |
+
+## Usage Example
+
+First, the watsonx.ai library is an optional dependency, so must be installed seperately:
+
+```
+pip install ibm-watsonx-ai
+```
+
+Optionally set environment variables (if not passing credentials to `create` directly):
+
+```sh
+export WATSONX_API_KEY="YOUR_WATSONX_API_KEY"
+export WATSONX_PROJECT_ID="YOUR_WATSONX_PROJECT_ID"
+```
+
+```python
+import os
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import EmbeddingFunctionRegistry
+
+watsonx_embed = EmbeddingFunctionRegistry
+  .get_instance()
+  .get("watsonx")
+  .create(
+    name="ibm/slate-125m-english-rtrvr",
+    # Uncomment and set these if not using environment variables
+    # api_key="your_api_key_here",
+    # project_id="your_project_id_here",
+    # url="your_watsonx_url_here",
+    # params={...},
+  )
+
+class TextModel(LanceModel):
+    text: str = watsonx_embed.SourceField()
+    vector: Vector(watsonx_embed.ndims()) = watsonx_embed.VectorField()
+
+data = [
+    {"text": "hello world"},
+    {"text": "goodbye world"},
+]
+
+db = lancedb.connect("~/.lancedb")
+tbl = db.create_table("watsonx_test", schema=TextModel, mode="overwrite")
+
+tbl.add(data)
+
+rs = tbl.search("hello").limit(1).to_pandas()
+print(rs)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/instructor_embedding.md

@@ -0,0 +1,50 @@
+# Instructor Embeddings
+[Instructor](https://instructor-embedding.github.io/) is an instruction-finetuned text embedding model that can generate text embeddings tailored to any task (e.g. classification, retrieval, clustering, text evaluation, etc.) and domains (e.g. science, finance, etc.) by simply providing the task instruction, without any finetuning.
+
+If you want to calculate customized embeddings for specific sentences, you can follow the unified template to write instructions.
+
+!!! info
+    Represent the `domain` `text_type` for `task_objective`:
+
+    * `domain` is optional, and it specifies the domain of the text, e.g. science, finance, medicine, etc.
+    * `text_type` is required, and it specifies the encoding unit, e.g. sentence, document, paragraph, etc.
+    * `task_objective` is optional, and it specifies the objective of embedding, e.g. retrieve a document, classify the sentence, etc.
+
+More information about the model can be found at the [source URL](https://github.com/xlang-ai/instructor-embedding).
+
+| Argument | Type | Default | Description |
+|---|---|---|---|
+| `name` | `str` | "hkunlp/instructor-base" | The name of the model to use |
+| `batch_size` | `int` | `32` | The batch size to use when generating embeddings |
+| `device` | `str` | `"cpu"` | The device to use when generating embeddings |
+| `show_progress_bar` | `bool` | `True` | Whether to show a progress bar when generating embeddings |
+| `normalize_embeddings` | `bool` | `True` | Whether to normalize the embeddings |
+| `quantize` | `bool` | `False` | Whether to quantize the model |
+| `source_instruction` | `str` | `"represent the docuement for retreival"` | The instruction for the source column |
+| `query_instruction` | `str` | `"represent the document for retreiving the most similar documents"` | The instruction for the query |
+
+
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry, InstuctorEmbeddingFunction
+
+instructor = get_registry().get("instructor").create(
+                            source_instruction="represent the docuement for retreival",
+                            query_instruction="represent the document for retreiving the most similar documents"
+                            )
+
+class Schema(LanceModel):
+    vector: Vector(instructor.ndims()) = instructor.VectorField()
+    text: str = instructor.SourceField()
+
+db = lancedb.connect("~/.lancedb")
+tbl = db.create_table("test", schema=Schema, mode="overwrite")
+
+texts = [{"text": "Capitalism has been dominant in the Western world since the end of feudalism, but most feel[who?] that..."},
+        {"text": "The disparate impact theory is especially controversial under the Fair Housing Act because the Act..."},
+        {"text": "Disparate impact in United States labor law refers to practices in employment, housing, and other areas that.."}]
+
+tbl.add(texts)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/jina_embedding.md

@@ -0,0 +1,39 @@
+# Jina Embeddings
+
+Jina embeddings are used to generate embeddings for text and image data.
+You also need to set the `JINA_API_KEY` environment variable to use the Jina API.
+
+You can find a list of supported models under [https://jina.ai/embeddings/](https://jina.ai/embeddings/)
+
+Supported parameters (to be passed in `create` method) are:
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `"jina-clip-v1"` | The model ID of the jina model to use |
+
+Usage Example:
+
+```python
+    import os
+    import lancedb
+    from lancedb.pydantic import LanceModel, Vector
+    from lancedb.embeddings import EmbeddingFunctionRegistry
+
+    os.environ['JINA_API_KEY'] = 'jina_*'
+
+    jina_embed = EmbeddingFunctionRegistry.get_instance().get("jina").create(name="jina-embeddings-v2-base-en")
+
+
+    class TextModel(LanceModel):
+        text: str = jina_embed.SourceField()
+        vector: Vector(jina_embed.ndims()) = jina_embed.VectorField()
+
+
+    data = [{"text": "hello world"},
+            {"text": "goodbye world"}]
+
+    db = lancedb.connect("~/.lancedb-2")
+    tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+    tbl.add(data)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/ollama_embedding.md

@@ -0,0 +1,37 @@
+# Ollama embeddings
+
+Generate embeddings via the [ollama](https://github.com/ollama/ollama-python) python library. More details:
+
+- [Ollama docs on embeddings](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embeddings)
+- [Ollama blog on embeddings](https://ollama.com/blog/embedding-models)
+
+| Parameter              | Type                       | Default Value            | Description                                                                                                                                    |
+|------------------------|----------------------------|--------------------------|------------------------------------------------------------------------------------------------------------------------------------------------|
+| `name`                 | `str`                      | `nomic-embed-text`       | The name of the model.                                                                                                                         |
+| `host`                 | `str`                      | `http://localhost:11434` | The Ollama host to connect to.                                                                                                                 |
+| `options`              | `ollama.Options` or `dict` | `None`                   | Additional model parameters listed in the documentation for the Modelfile such as `temperature`. |
+| `keep_alive`           | `float` or `str`           | `"5m"`                   | Controls how long the model will stay loaded into memory following the request.                                                                |
+| `ollama_client_kwargs` | `dict`                     | `{}`                     | kwargs that can be past to the `ollama.Client`.                                                                                                |
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+db = lancedb.connect("/tmp/db")
+func = get_registry().get("ollama").create(name="nomic-embed-text")
+
+class Words(LanceModel):
+    text: str = func.SourceField()
+    vector: Vector(func.ndims()) = func.VectorField()
+
+table = db.create_table("words", schema=Words, mode="overwrite")
+table.add([
+    {"text": "hello world"},
+    {"text": "goodbye world"}
+])
+
+query = "greetings"
+actual = table.search(query).limit(1).to_pydantic(Words)[0]
+print(actual.text)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/openai_embedding.md

@@ -0,0 +1,35 @@
+# OpenAI embeddings
+
+LanceDB registers the OpenAI embeddings function in the registry by default, as `openai`. Below are the parameters that you can customize when creating the instances:
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `"text-embedding-ada-002"` | The name of the model. |
+| `dim` | `int` |  Model default   | For OpenAI's newer text-embedding-3 model, we can specify a dimensionality that is smaller than the 1536 size. This feature supports it |
+| `use_azure` | bool | `False` | Set true to use Azure OpenAPI SDK |
+
+
+```python
+import lancedb
+from lancedb.pydantic import LanceModel, Vector
+from lancedb.embeddings import get_registry
+
+db = lancedb.connect("/tmp/db")
+func = get_registry().get("openai").create(name="text-embedding-ada-002")
+
+class Words(LanceModel):
+    text: str = func.SourceField()
+    vector: Vector(func.ndims()) = func.VectorField()
+
+table = db.create_table("words", schema=Words, mode="overwrite")
+table.add(
+    [
+        {"text": "hello world"},
+        {"text": "goodbye world"}
+    ]
+    )
+
+query = "greetings"
+actual = table.search(query).limit(1).to_pydantic(Words)[0]
+print(actual.text)
+```

docs/src/embeddings/available_embedding_models/text_embedding_functions/sentence_transformers.md

@@ -0,0 +1,174 @@
+# Sentence transformers
+Allows you to set parameters when registering a `sentence-transformers` object.
+
+!!! info
+    Sentence transformer embeddings are normalized by default. It is recommended to use normalized embeddings for similarity search.
+
+| Parameter | Type | Default Value | Description |
+|---|---|---|---|
+| `name` | `str` | `all-MiniLM-L6-v2` | The name of the model |
+| `device` | `str` | `cpu` | The device to run the model on (can be `cpu` or `gpu`) |
+| `normalize` | `bool` | `True` | Whether to normalize the input text before feeding it to the model |
+| `trust_remote_code` | `bool` | `False` | Whether to trust and execute remote code from the model's Huggingface repository |
+
+
+??? "Check out available sentence-transformer models here!"
+    ```markdown
+    - sentence-transformers/all-MiniLM-L12-v2
+    - sentence-transformers/paraphrase-mpnet-base-v2
+    - sentence-transformers/gtr-t5-base
+    - sentence-transformers/LaBSE
+    - sentence-transformers/all-MiniLM-L6-v2
+    - sentence-transformers/bert-base-nli-max-tokens
+    - sentence-transformers/bert-base-nli-mean-tokens
+    - sentence-transformers/bert-base-nli-stsb-mean-tokens
+    - sentence-transformers/bert-base-wikipedia-sections-mean-tokens
+    - sentence-transformers/bert-large-nli-cls-token
+    - sentence-transformers/bert-large-nli-max-tokens
+    - sentence-transformers/bert-large-nli-mean-tokens
+    - sentence-transformers/bert-large-nli-stsb-mean-tokens
+    - sentence-transformers/distilbert-base-nli-max-tokens
+    - sentence-transformers/distilbert-base-nli-mean-tokens
+    - sentence-transformers/distilbert-base-nli-stsb-mean-tokens
+    - sentence-transformers/distilroberta-base-msmarco-v1
+    - sentence-transformers/distilroberta-base-msmarco-v2
+    - sentence-transformers/nli-bert-base-cls-pooling
+    - sentence-transformers/nli-bert-base-max-pooling
+    - sentence-transformers/nli-bert-base
+    - sentence-transformers/nli-bert-large-cls-pooling
+    - sentence-transformers/nli-bert-large-max-pooling
+    - sentence-transformers/nli-bert-large
+    - sentence-transformers/nli-distilbert-base-max-pooling
+    - sentence-transformers/nli-distilbert-base
+    - sentence-transformers/nli-roberta-base
+    - sentence-transformers/nli-roberta-large
+    - sentence-transformers/roberta-base-nli-mean-tokens
+    - sentence-transformers/roberta-base-nli-stsb-mean-tokens
+    - sentence-transformers/roberta-large-nli-mean-tokens
+    - sentence-transformers/roberta-large-nli-stsb-mean-tokens
+    - sentence-transformers/stsb-bert-base
+    - sentence-transformers/stsb-bert-large
+    - sentence-transformers/stsb-distilbert-base
+    - sentence-transformers/stsb-roberta-base
+    - sentence-transformers/stsb-roberta-large
+    - sentence-transformers/xlm-r-100langs-bert-base-nli-mean-tokens
+    - sentence-transformers/xlm-r-100langs-bert-base-nli-stsb-mean-tokens
+    - sentence-transformers/xlm-r-base-en-ko-nli-ststb
+    - sentence-transformers/xlm-r-bert-base-nli-mean-tokens
+    - sentence-transformers/xlm-r-bert-base-nli-stsb-mean-tokens
+    - sentence-transformers/xlm-r-large-en-ko-nli-ststb
+    - sentence-transformers/bert-base-nli-cls-token
+    - sentence-transformers/all-distilroberta-v1
+    - sentence-transformers/multi-qa-MiniLM-L6-dot-v1
+    - sentence-transformers/multi-qa-distilbert-cos-v1
+    - sentence-transformers/multi-qa-distilbert-dot-v1
+    - sentence-transformers/multi-qa-mpnet-base-cos-v1
+    - sentence-transformers/multi-qa-mpnet-base-dot-v1
+    - sentence-transformers/nli-distilroberta-base-v2
+    - sentence-transformers/all-MiniLM-L6-v1
+    - sentence-transformers/all-mpnet-base-v1
+    - sentence-transformers/all-mpnet-base-v2
+    - sentence-transformers/all-roberta-large-v1
+    - sentence-transformers/allenai-specter
+    - sentence-transformers/average_word_embeddings_glove.6B.300d
+    - sentence-transformers/average_word_embeddings_glove.840B.300d
+    - sentence-transformers/average_word_embeddings_komninos
+    - sentence-transformers/average_word_embeddings_levy_dependency
+    - sentence-transformers/clip-ViT-B-32-multilingual-v1
+    - sentence-transformers/clip-ViT-B-32
+    - sentence-transformers/distilbert-base-nli-stsb-quora-ranking
+    - sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking
+    - sentence-transformers/distilroberta-base-paraphrase-v1
+    - sentence-transformers/distiluse-base-multilingual-cased-v1
+    - sentence-transformers/distiluse-base-multilingual-cased-v2
+    - sentence-transformers/distiluse-base-multilingual-cased
+    - sentence-transformers/facebook-dpr-ctx_encoder-multiset-base
+    - sentence-transformers/facebook-dpr-ctx_encoder-single-nq-base
+    - sentence-transformers/facebook-dpr-question_encoder-multiset-base
+    - sentence-transformers/facebook-dpr-question_encoder-single-nq-base
+    - sentence-transformers/gtr-t5-large
+    - sentence-transformers/gtr-t5-xl
+    - sentence-transformers/gtr-t5-xxl
+    - sentence-transformers/msmarco-MiniLM-L-12-v3
+    - sentence-transformers/msmarco-MiniLM-L-6-v3
+    - sentence-transformers/msmarco-MiniLM-L12-cos-v5
+    - sentence-transformers/msmarco-MiniLM-L6-cos-v5
+    - sentence-transformers/msmarco-bert-base-dot-v5
+    - sentence-transformers/msmarco-bert-co-condensor
+    - sentence-transformers/msmarco-distilbert-base-dot-prod-v3
+    - sentence-transformers/msmarco-distilbert-base-tas-b
+    - sentence-transformers/msmarco-distilbert-base-v2
+    - sentence-transformers/msmarco-distilbert-base-v3
+    - sentence-transformers/msmarco-distilbert-base-v4
+    - sentence-transformers/msmarco-distilbert-cos-v5
+    - sentence-transformers/msmarco-distilbert-dot-v5
+    - sentence-transformers/msmarco-distilbert-multilingual-en-de-v2-tmp-lng-aligned
+    - sentence-transformers/msmarco-distilbert-multilingual-en-de-v2-tmp-trained-scratch
+    - sentence-transformers/msmarco-distilroberta-base-v2
+    - sentence-transformers/msmarco-roberta-base-ance-firstp
+    - sentence-transformers/msmarco-roberta-base-v2
+    - sentence-transformers/msmarco-roberta-base-v3
+    - sentence-transformers/multi-qa-MiniLM-L6-cos-v1
+    - sentence-transformers/nli-mpnet-base-v2
+    - sentence-transformers/nli-roberta-base-v2
+    - sentence-transformers/nq-distilbert-base-v1
+    - sentence-transformers/paraphrase-MiniLM-L12-v2
+    - sentence-transformers/paraphrase-MiniLM-L3-v2
+    - sentence-transformers/paraphrase-MiniLM-L6-v2
+    - sentence-transformers/paraphrase-TinyBERT-L6-v2
+    - sentence-transformers/paraphrase-albert-base-v2
+    - sentence-transformers/paraphrase-albert-small-v2
+    - sentence-transformers/paraphrase-distilroberta-base-v1
+    - sentence-transformers/paraphrase-distilroberta-base-v2
+    - sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
+    - sentence-transformers/paraphrase-multilingual-mpnet-base-v2
+    - sentence-transformers/paraphrase-xlm-r-multilingual-v1
+    - sentence-transformers/quora-distilbert-base
+    - sentence-transformers/quora-distilbert-multilingual
+    - sentence-transformers/sentence-t5-base
+    - sentence-transformers/sentence-t5-large
+    - sentence-transformers/sentence-t5-xxl
+    - sentence-transformers/sentence-t5-xl
+    - sentence-transformers/stsb-distilroberta-base-v2
+    - sentence-transformers/stsb-mpnet-base-v2
+    - sentence-transformers/stsb-roberta-base-v2
+    - sentence-transformers/stsb-xlm-r-multilingual
+    - sentence-transformers/xlm-r-distilroberta-base-paraphrase-v1
+    - sentence-transformers/clip-ViT-L-14
+    - sentence-transformers/clip-ViT-B-16
+    - sentence-transformers/use-cmlm-multilingual
+    - sentence-transformers/all-MiniLM-L12-v1
+    ```
+
+!!! info
+    You can also load many other model architectures from the library. For example models from sources such as BAAI, nomic, salesforce research, etc.
+    See this HF hub page for all [supported models](https://huggingface.co/models?library=sentence-transformers).
+
+!!! note "BAAI Embeddings example"
+    Here is an example that uses BAAI embedding model from the HuggingFace Hub [supported models](https://huggingface.co/models?library=sentence-transformers)
+    ```python
+    import lancedb
+    from lancedb.pydantic import LanceModel, Vector
+    from lancedb.embeddings import get_registry
+
+    db = lancedb.connect("/tmp/db")
+    model = get_registry().get("sentence-transformers").create(name="BAAI/bge-small-en-v1.5", device="cpu")
+
+    class Words(LanceModel):
+        text: str = model.SourceField()
+        vector: Vector(model.ndims()) = model.VectorField()
+
+    table = db.create_table("words", schema=Words)
+    table.add(
+        [
+            {"text": "hello world"},
+            {"text": "goodbye world"}
+        ]
+    )
+
+    query = "greetings"
+    actual = table.search(query).limit(1).to_pydantic(Words)[0]
+    print(actual.text)
+    ```
+Visit sentence-transformers [HuggingFace HUB](https://huggingface.co/sentence-transformers) page for more information on the available models.
+

docs/src/embeddings/available_embedding_models/text_embedding_functions/voyageai_embedding.md

@@ -0,0 +1,51 @@
+# VoyageAI Embeddings
+
+Voyage AI provides cutting-edge embedding and rerankers.
+
+
+Using voyageai API requires voyageai package, which can be installed using `pip install voyageai`. Voyage AI embeddings are used to generate embeddings for text data. The embeddings can be used for various tasks like semantic search, clustering, and classification.
+You also need to set the `VOYAGE_API_KEY` environment variable to use the VoyageAI API.
+
+Supported models are:
+
+- voyage-3
+- voyage-3-lite
+- voyage-finance-2
+- voyage-multilingual-2
+- voyage-law-2
+- voyage-code-2
+
+
+Supported parameters (to be passed in `create` method) are:
+
+| Parameter | Type | Default Value | Description |
+|---|---|--------|---------|
+| `name` | `str` | `None` | The model ID of the model to use. Supported base models for Text Embeddings: voyage-3, voyage-3-lite, voyage-finance-2, voyage-multilingual-2, voyage-law-2, voyage-code-2 |
+| `input_type` | `str` | `None` | Type of the input text. Default to None. Other options: query, document. |
+| `truncation` | `bool` | `True` | Whether to truncate the input texts to fit within the context length. |
+
+
+Usage Example:
+    
+```python
+    import lancedb
+    from lancedb.pydantic import LanceModel, Vector
+    from lancedb.embeddings import EmbeddingFunctionRegistry
+
+    voyageai = EmbeddingFunctionRegistry
+        .get_instance()
+        .get("voyageai")
+        .create(name="voyage-3")
+
+    class TextModel(LanceModel):
+        text: str = voyageai.SourceField()
+        vector: Vector(voyageai.ndims()) =  voyageai.VectorField()
+
+    data = [ { "text": "hello world" },
+            { "text": "goodbye world" }]
+
+    db = lancedb.connect("~/.lancedb")
+    tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+
+    tbl.add(data)
+```

docs/src/embeddings/custom_embedding_function.md

@@ -15,198 +15,234 @@ There is another optional layer of abstraction available: `TextEmbeddingFunction
 
 Let's implement `SentenceTransformerEmbeddings` class. All you need to do is implement the `generate_embeddings()` and `ndims` function to handle the input types you expect and register the class in the global `EmbeddingFunctionRegistry`
 
-```python
-from lancedb.embeddings import register
-from lancedb.util import attempt_import_or_raise
 
-@register("sentence-transformers")
-class SentenceTransformerEmbeddings(TextEmbeddingFunction):
-    name: str = "all-MiniLM-L6-v2"
-    # set more default instance vars like device, etc.
+=== "Python"
 
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-        self._ndims = None
-    
-    def generate_embeddings(self, texts):
-        return self._embedding_model().encode(list(texts), ...).tolist()
+    ```python
+    from lancedb.embeddings import register
+    from lancedb.util import attempt_import_or_raise
 
-    def ndims(self):
-        if self._ndims is None:
-            self._ndims = len(self.generate_embeddings("foo")[0])
-        return self._ndims
+    @register("sentence-transformers")
+    class SentenceTransformerEmbeddings(TextEmbeddingFunction):
+        name: str = "all-MiniLM-L6-v2"
+        # set more default instance vars like device, etc.
 
-    @cached(cache={}) 
-    def _embedding_model(self):
-        return sentence_transformers.SentenceTransformer(name)
-```
+        def __init__(self, **kwargs):
+            super().__init__(**kwargs)
+            self._ndims = None
 
-This is a stripped down version of our implementation of `SentenceTransformerEmbeddings` that removes certain optimizations and defaul settings.
+        def generate_embeddings(self, texts):
+            return self._embedding_model().encode(list(texts), ...).tolist()
+
+        def ndims(self):
+            if self._ndims is None:
+                self._ndims = len(self.generate_embeddings("foo")[0])
+            return self._ndims
+
+        @cached(cache={})
+        def _embedding_model(self):
+            return sentence_transformers.SentenceTransformer(name)
+    ```
+
+=== "TypeScript"
+
+    ```ts
+    --8<--- "nodejs/examples/custom_embedding_function.test.ts:imports"
+
+    --8<--- "nodejs/examples/custom_embedding_function.test.ts:embedding_impl"
+    ```
+
+
+This is a stripped down version of our implementation of `SentenceTransformerEmbeddings` that removes certain optimizations and default settings.
+
+!!! danger "Use sensitive keys to prevent leaking secrets"
+    To prevent leaking secrets, such as API keys, you should add any sensitive
+    parameters of an embedding function to the output of the
+    [sensitive_keys()][lancedb.embeddings.base.EmbeddingFunction.sensitive_keys] /
+    [getSensitiveKeys()](../../js/namespaces/embedding/classes/EmbeddingFunction/#getsensitivekeys)
+    method. This prevents users from accidentally instantiating the embedding
+    function with hard-coded secrets.
 
 Now you can use this embedding function to create your table schema and that's it! you can then ingest data and run queries without manually vectorizing the inputs.
 
-```python
-from lancedb.pydantic import LanceModel, Vector
+=== "Python"
 
-registry = EmbeddingFunctionRegistry.get_instance()
-stransformer = registry.get("sentence-transformers").create()
+    ```python
+    from lancedb.pydantic import LanceModel, Vector
 
-class TextModelSchema(LanceModel):
-    vector: Vector(stransformer.ndims) = stransformer.VectorField()
-    text: str = stransformer.SourceField()
+    registry = EmbeddingFunctionRegistry.get_instance()
+    stransformer = registry.get("sentence-transformers").create()
 
-tbl = db.create_table("table", schema=TextModelSchema)
+    class TextModelSchema(LanceModel):
+        vector: Vector(stransformer.ndims) = stransformer.VectorField()
+        text: str = stransformer.SourceField()
 
-tbl.add(pd.DataFrame({"text": ["halo", "world"]}))
-result = tbl.search("world").limit(5)
-```
+    tbl = db.create_table("table", schema=TextModelSchema)
 
-NOTE:
+    tbl.add(pd.DataFrame({"text": ["halo", "world"]}))
+    result = tbl.search("world").limit(5)
+    ```
 
-You can always implement the `EmbeddingFunction` interface directly if you want or need to, `TextEmbeddingFunction` just makes it much simpler and faster for you to do so, by setting up the boiler plat for text-specific use case
+=== "TypeScript"
+
+    ```ts
+    --8<--- "nodejs/examples/custom_embedding_function.test.ts:call_custom_function"
+    ```
+
+!!! note
+
+    You can always implement the `EmbeddingFunction` interface directly if you want or need to, `TextEmbeddingFunction` just makes it much simpler and faster for you to do so, by setting up the boiler plat for text-specific use case
 
 ## Multi-modal embedding function example
-You can also use the `EmbeddingFunction` interface to implement more complex workflows such as multi-modal embedding function support. LanceDB implements `OpenClipEmeddingFunction` class that suppports multi-modal seach. Here's the implementation that you can use as a reference to build your own multi-modal embedding functions.
+You can also use the `EmbeddingFunction` interface to implement more complex workflows such as multi-modal embedding function support.
 
-```python
-@register("open-clip")
-class OpenClipEmbeddings(EmbeddingFunction):
-    name: str = "ViT-B-32"
-    pretrained: str = "laion2b_s34b_b79k"
-    device: str = "cpu"
-    batch_size: int = 64
-    normalize: bool = True
-    _model = PrivateAttr()
-    _preprocess = PrivateAttr()
-    _tokenizer = PrivateAttr()
+=== "Python"
 
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        open_clip = attempt_import_or_raise("open_clip", "open-clip") # EmbeddingFunction util to import external libs and raise if not found
-        model, _, preprocess = open_clip.create_model_and_transforms(
-            self.name, pretrained=self.pretrained
-        )
-        model.to(self.device)
-        self._model, self._preprocess = model, preprocess
-        self._tokenizer = open_clip.get_tokenizer(self.name)
-        self._ndims = None
+    LanceDB implements `OpenClipEmeddingFunction` class that suppports multi-modal seach. Here's the implementation that you can use as a reference to build your own multi-modal embedding functions.
 
-    def ndims(self):
-        if self._ndims is None:
-            self._ndims = self.generate_text_embeddings("foo").shape[0]
-        return self._ndims
+    ```python
+    @register("open-clip")
+    class OpenClipEmbeddings(EmbeddingFunction):
+        name: str = "ViT-B-32"
+        pretrained: str = "laion2b_s34b_b79k"
+        device: str = "cpu"
+        batch_size: int = 64
+        normalize: bool = True
+        _model = PrivateAttr()
+        _preprocess = PrivateAttr()
+        _tokenizer = PrivateAttr()
 
-    def compute_query_embeddings(
-        self, query: Union[str, "PIL.Image.Image"], *args, **kwargs
-    ) -> List[np.ndarray]:
-        """
-        Compute the embeddings for a given user query
+        def __init__(self, *args, **kwargs):
+            super().__init__(*args, **kwargs)
+            open_clip = attempt_import_or_raise("open_clip", "open-clip") # EmbeddingFunction util to import external libs and raise if not found
+            model, _, preprocess = open_clip.create_model_and_transforms(
+                self.name, pretrained=self.pretrained
+            )
+            model.to(self.device)
+            self._model, self._preprocess = model, preprocess
+            self._tokenizer = open_clip.get_tokenizer(self.name)
+            self._ndims = None
 
-        Parameters
-        ----------
-        query : Union[str, PIL.Image.Image]
-            The query to embed. A query can be either text or an image.
-        """
-        if isinstance(query, str):
-            return [self.generate_text_embeddings(query)]
-        else:
+        def ndims(self):
+            if self._ndims is None:
+                self._ndims = self.generate_text_embeddings("foo").shape[0]
+            return self._ndims
+
+        def compute_query_embeddings(
+            self, query: Union[str, "PIL.Image.Image"], *args, **kwargs
+        ) -> List[np.ndarray]:
+            """
+            Compute the embeddings for a given user query
+
+            Parameters
+            ----------
+            query : Union[str, PIL.Image.Image]
+                The query to embed. A query can be either text or an image.
+            """
+            if isinstance(query, str):
+                return [self.generate_text_embeddings(query)]
+            else:
+                PIL = attempt_import_or_raise("PIL", "pillow")
+                if isinstance(query, PIL.Image.Image):
+                    return [self.generate_image_embedding(query)]
+                else:
+                    raise TypeError("OpenClip supports str or PIL Image as query")
+
+        def generate_text_embeddings(self, text: str) -> np.ndarray:
+            torch = attempt_import_or_raise("torch")
+            text = self.sanitize_input(text)
+            text = self._tokenizer(text)
+            text.to(self.device)
+            with torch.no_grad():
+                text_features = self._model.encode_text(text.to(self.device))
+                if self.normalize:
+                    text_features /= text_features.norm(dim=-1, keepdim=True)
+                return text_features.cpu().numpy().squeeze()
+
+        def sanitize_input(self, images: IMAGES) -> Union[List[bytes], np.ndarray]:
+            """
+            Sanitize the input to the embedding function.
+            """
+            if isinstance(images, (str, bytes)):
+                images = [images]
+            elif isinstance(images, pa.Array):
+                images = images.to_pylist()
+            elif isinstance(images, pa.ChunkedArray):
+                images = images.combine_chunks().to_pylist()
+            return images
+
+        def compute_source_embeddings(
+            self, images: IMAGES, *args, **kwargs
+        ) -> List[np.array]:
+            """
+            Get the embeddings for the given images
+            """
+            images = self.sanitize_input(images)
+            embeddings = []
+            for i in range(0, len(images), self.batch_size):
+                j = min(i + self.batch_size, len(images))
+                batch = images[i:j]
+                embeddings.extend(self._parallel_get(batch))
+            return embeddings
+
+        def _parallel_get(self, images: Union[List[str], List[bytes]]) -> List[np.ndarray]:
+            """
+            Issue concurrent requests to retrieve the image data
+            """
+            with concurrent.futures.ThreadPoolExecutor() as executor:
+                futures = [
+                    executor.submit(self.generate_image_embedding, image)
+                    for image in images
+                ]
+                return [future.result() for future in futures]
+
+        def generate_image_embedding(
+            self, image: Union[str, bytes, "PIL.Image.Image"]
+        ) -> np.ndarray:
+            """
+            Generate the embedding for a single image
+
+            Parameters
+            ----------
+            image : Union[str, bytes, PIL.Image.Image]
+                The image to embed. If the image is a str, it is treated as a uri.
+                If the image is bytes, it is treated as the raw image bytes.
+            """
+            torch = attempt_import_or_raise("torch")
+            # TODO handle retry and errors for https
+            image = self._to_pil(image)
+            image = self._preprocess(image).unsqueeze(0)
+            with torch.no_grad():
+                return self._encode_and_normalize_image(image)
+
+        def _to_pil(self, image: Union[str, bytes]):
             PIL = attempt_import_or_raise("PIL", "pillow")
-            if isinstance(query, PIL.Image.Image):
-                return [self.generate_image_embedding(query)]
-            else:
-                raise TypeError("OpenClip supports str or PIL Image as query")
+            if isinstance(image, bytes):
+                return PIL.Image.open(io.BytesIO(image))
+            if isinstance(image, PIL.Image.Image):
+                return image
+            elif isinstance(image, str):
+                parsed = urlparse.urlparse(image)
+                # TODO handle drive letter on windows.
+                if parsed.scheme == "file":
+                    return PIL.Image.open(parsed.path)
+                elif parsed.scheme == "":
+                    return PIL.Image.open(image if os.name == "nt" else parsed.path)
+                elif parsed.scheme.startswith("http"):
+                    return PIL.Image.open(io.BytesIO(url_retrieve(image)))
+                else:
+                    raise NotImplementedError("Only local and http(s) urls are supported")
 
-    def generate_text_embeddings(self, text: str) -> np.ndarray:
-        torch = attempt_import_or_raise("torch")
-        text = self.sanitize_input(text)
-        text = self._tokenizer(text)
-        text.to(self.device)
-        with torch.no_grad():
-            text_features = self._model.encode_text(text.to(self.device))
+        def _encode_and_normalize_image(self, image_tensor: "torch.Tensor"):
+            """
+            encode a single image tensor and optionally normalize the output
+            """
+            image_features = self._model.encode_image(image_tensor)
             if self.normalize:
-                text_features /= text_features.norm(dim=-1, keepdim=True)
-            return text_features.cpu().numpy().squeeze()
+                image_features /= image_features.norm(dim=-1, keepdim=True)
+            return image_features.cpu().numpy().squeeze()
+    ```
 
-    def sanitize_input(self, images: IMAGES) -> Union[List[bytes], np.ndarray]:
-        """
-        Sanitize the input to the embedding function.
-        """
-        if isinstance(images, (str, bytes)):
-            images = [images]
-        elif isinstance(images, pa.Array):
-            images = images.to_pylist()
-        elif isinstance(images, pa.ChunkedArray):
-            images = images.combine_chunks().to_pylist()
-        return images
+=== "TypeScript"
 
-    def compute_source_embeddings(
-        self, images: IMAGES, *args, **kwargs
-    ) -> List[np.array]:
-        """
-        Get the embeddings for the given images
-        """
-        images = self.sanitize_input(images)
-        embeddings = []
-        for i in range(0, len(images), self.batch_size):
-            j = min(i + self.batch_size, len(images))
-            batch = images[i:j]
-            embeddings.extend(self._parallel_get(batch))
-        return embeddings
-
-    def _parallel_get(self, images: Union[List[str], List[bytes]]) -> List[np.ndarray]:
-        """
-        Issue concurrent requests to retrieve the image data
-        """
-        with concurrent.futures.ThreadPoolExecutor() as executor:
-            futures = [
-                executor.submit(self.generate_image_embedding, image)
-                for image in images
-            ]
-            return [future.result() for future in futures]
-
-    def generate_image_embedding(
-        self, image: Union[str, bytes, "PIL.Image.Image"]
-    ) -> np.ndarray:
-        """
-        Generate the embedding for a single image
-
-        Parameters
-        ----------
-        image : Union[str, bytes, PIL.Image.Image]
-            The image to embed. If the image is a str, it is treated as a uri.
-            If the image is bytes, it is treated as the raw image bytes.
-        """
-        torch = attempt_import_or_raise("torch")
-        # TODO handle retry and errors for https
-        image = self._to_pil(image)
-        image = self._preprocess(image).unsqueeze(0)
-        with torch.no_grad():
-            return self._encode_and_normalize_image(image)
-
-    def _to_pil(self, image: Union[str, bytes]):
-        PIL = attempt_import_or_raise("PIL", "pillow")
-        if isinstance(image, bytes):
-            return PIL.Image.open(io.BytesIO(image))
-        if isinstance(image, PIL.Image.Image):
-            return image
-        elif isinstance(image, str):
-            parsed = urlparse.urlparse(image)
-            # TODO handle drive letter on windows.
-            if parsed.scheme == "file":
-                return PIL.Image.open(parsed.path)
-            elif parsed.scheme == "":
-                return PIL.Image.open(image if os.name == "nt" else parsed.path)
-            elif parsed.scheme.startswith("http"):
-                return PIL.Image.open(io.BytesIO(url_retrieve(image)))
-            else:
-                raise NotImplementedError("Only local and http(s) urls are supported")
-
-    def _encode_and_normalize_image(self, image_tensor: "torch.Tensor"):
-        """
-        encode a single image tensor and optionally normalize the output
-        """
-        image_features = self._model.encode_image(image_tensor)
-        if self.normalize:
-            image_features /= image_features.norm(dim=-1, keepdim=True)
-        return image_features.cpu().numpy().squeeze()
-```
+    Coming Soon! See this [issue](https://github.com/lancedb/lancedb/issues/1482) to track the status!

docs/src/embeddings/default_embedding_functions.md

@@ -1,722 +1,86 @@
-There are various embedding functions available out of the box with LanceDB to manage your embeddings implicitly. We're actively working on adding other popular embedding APIs and models.
+# 📚 Available Embedding Models
 
-## Text embedding functions
-Contains the text embedding functions registered by default.
+There are various embedding functions available out of the box with LanceDB to manage your embeddings implicitly. We're actively working on adding other popular embedding APIs and models. 🚀
 
-* Embedding functions have an inbuilt rate limit handler wrapper for source and query embedding function calls that retry with exponential backoff. 
-* Each `EmbeddingFunction` implementation automatically takes `max_retries` as an argument which has the default value of 7.
+Before jumping on the list of available models, let's understand how to get an embedding model initialized and configured to use in our code: 
 
-### Sentence transformers
-Allows you to set parameters when registering a `sentence-transformers` object.
-
-!!! info
-    Sentence transformer embeddings are normalized by default. It is recommended to use normalized embeddings for similarity search.
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `all-MiniLM-L6-v2` | The name of the model |
-| `device` | `str` | `cpu` | The device to run the model on (can be `cpu` or `gpu`) |
-| `normalize` | `bool` | `True` | Whether to normalize the input text before feeding it to the model |
-
-
-??? "Check out available sentence-transformer models here!"
-    ```markdown
-    - sentence-transformers/all-MiniLM-L12-v2
-    - sentence-transformers/paraphrase-mpnet-base-v2
-    - sentence-transformers/gtr-t5-base
-    - sentence-transformers/LaBSE
-    - sentence-transformers/all-MiniLM-L6-v2
-    - sentence-transformers/bert-base-nli-max-tokens
-    - sentence-transformers/bert-base-nli-mean-tokens
-    - sentence-transformers/bert-base-nli-stsb-mean-tokens
-    - sentence-transformers/bert-base-wikipedia-sections-mean-tokens
-    - sentence-transformers/bert-large-nli-cls-token
-    - sentence-transformers/bert-large-nli-max-tokens
-    - sentence-transformers/bert-large-nli-mean-tokens
-    - sentence-transformers/bert-large-nli-stsb-mean-tokens
-    - sentence-transformers/distilbert-base-nli-max-tokens
-    - sentence-transformers/distilbert-base-nli-mean-tokens
-    - sentence-transformers/distilbert-base-nli-stsb-mean-tokens
-    - sentence-transformers/distilroberta-base-msmarco-v1
-    - sentence-transformers/distilroberta-base-msmarco-v2
-    - sentence-transformers/nli-bert-base-cls-pooling
-    - sentence-transformers/nli-bert-base-max-pooling
-    - sentence-transformers/nli-bert-base
-    - sentence-transformers/nli-bert-large-cls-pooling
-    - sentence-transformers/nli-bert-large-max-pooling
-    - sentence-transformers/nli-bert-large
-    - sentence-transformers/nli-distilbert-base-max-pooling
-    - sentence-transformers/nli-distilbert-base
-    - sentence-transformers/nli-roberta-base
-    - sentence-transformers/nli-roberta-large
-    - sentence-transformers/roberta-base-nli-mean-tokens
-    - sentence-transformers/roberta-base-nli-stsb-mean-tokens
-    - sentence-transformers/roberta-large-nli-mean-tokens
-    - sentence-transformers/roberta-large-nli-stsb-mean-tokens
-    - sentence-transformers/stsb-bert-base
-    - sentence-transformers/stsb-bert-large
-    - sentence-transformers/stsb-distilbert-base
-    - sentence-transformers/stsb-roberta-base
-    - sentence-transformers/stsb-roberta-large
-    - sentence-transformers/xlm-r-100langs-bert-base-nli-mean-tokens
-    - sentence-transformers/xlm-r-100langs-bert-base-nli-stsb-mean-tokens
-    - sentence-transformers/xlm-r-base-en-ko-nli-ststb
-    - sentence-transformers/xlm-r-bert-base-nli-mean-tokens
-    - sentence-transformers/xlm-r-bert-base-nli-stsb-mean-tokens
-    - sentence-transformers/xlm-r-large-en-ko-nli-ststb
-    - sentence-transformers/bert-base-nli-cls-token
-    - sentence-transformers/all-distilroberta-v1
-    - sentence-transformers/multi-qa-MiniLM-L6-dot-v1
-    - sentence-transformers/multi-qa-distilbert-cos-v1
-    - sentence-transformers/multi-qa-distilbert-dot-v1
-    - sentence-transformers/multi-qa-mpnet-base-cos-v1
-    - sentence-transformers/multi-qa-mpnet-base-dot-v1
-    - sentence-transformers/nli-distilroberta-base-v2
-    - sentence-transformers/all-MiniLM-L6-v1
-    - sentence-transformers/all-mpnet-base-v1
-    - sentence-transformers/all-mpnet-base-v2
-    - sentence-transformers/all-roberta-large-v1
-    - sentence-transformers/allenai-specter
-    - sentence-transformers/average_word_embeddings_glove.6B.300d
-    - sentence-transformers/average_word_embeddings_glove.840B.300d
-    - sentence-transformers/average_word_embeddings_komninos
-    - sentence-transformers/average_word_embeddings_levy_dependency
-    - sentence-transformers/clip-ViT-B-32-multilingual-v1
-    - sentence-transformers/clip-ViT-B-32
-    - sentence-transformers/distilbert-base-nli-stsb-quora-ranking
-    - sentence-transformers/distilbert-multilingual-nli-stsb-quora-ranking
-    - sentence-transformers/distilroberta-base-paraphrase-v1
-    - sentence-transformers/distiluse-base-multilingual-cased-v1
-    - sentence-transformers/distiluse-base-multilingual-cased-v2
-    - sentence-transformers/distiluse-base-multilingual-cased
-    - sentence-transformers/facebook-dpr-ctx_encoder-multiset-base
-    - sentence-transformers/facebook-dpr-ctx_encoder-single-nq-base
-    - sentence-transformers/facebook-dpr-question_encoder-multiset-base
-    - sentence-transformers/facebook-dpr-question_encoder-single-nq-base
-    - sentence-transformers/gtr-t5-large
-    - sentence-transformers/gtr-t5-xl
-    - sentence-transformers/gtr-t5-xxl
-    - sentence-transformers/msmarco-MiniLM-L-12-v3
-    - sentence-transformers/msmarco-MiniLM-L-6-v3
-    - sentence-transformers/msmarco-MiniLM-L12-cos-v5
-    - sentence-transformers/msmarco-MiniLM-L6-cos-v5
-    - sentence-transformers/msmarco-bert-base-dot-v5
-    - sentence-transformers/msmarco-bert-co-condensor
-    - sentence-transformers/msmarco-distilbert-base-dot-prod-v3
-    - sentence-transformers/msmarco-distilbert-base-tas-b
-    - sentence-transformers/msmarco-distilbert-base-v2
-    - sentence-transformers/msmarco-distilbert-base-v3
-    - sentence-transformers/msmarco-distilbert-base-v4
-    - sentence-transformers/msmarco-distilbert-cos-v5
-    - sentence-transformers/msmarco-distilbert-dot-v5
-    - sentence-transformers/msmarco-distilbert-multilingual-en-de-v2-tmp-lng-aligned
-    - sentence-transformers/msmarco-distilbert-multilingual-en-de-v2-tmp-trained-scratch
-    - sentence-transformers/msmarco-distilroberta-base-v2
-    - sentence-transformers/msmarco-roberta-base-ance-firstp
-    - sentence-transformers/msmarco-roberta-base-v2
-    - sentence-transformers/msmarco-roberta-base-v3
-    - sentence-transformers/multi-qa-MiniLM-L6-cos-v1
-    - sentence-transformers/nli-mpnet-base-v2
-    - sentence-transformers/nli-roberta-base-v2
-    - sentence-transformers/nq-distilbert-base-v1
-    - sentence-transformers/paraphrase-MiniLM-L12-v2
-    - sentence-transformers/paraphrase-MiniLM-L3-v2
-    - sentence-transformers/paraphrase-MiniLM-L6-v2
-    - sentence-transformers/paraphrase-TinyBERT-L6-v2
-    - sentence-transformers/paraphrase-albert-base-v2
-    - sentence-transformers/paraphrase-albert-small-v2
-    - sentence-transformers/paraphrase-distilroberta-base-v1
-    - sentence-transformers/paraphrase-distilroberta-base-v2
-    - sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
-    - sentence-transformers/paraphrase-multilingual-mpnet-base-v2
-    - sentence-transformers/paraphrase-xlm-r-multilingual-v1
-    - sentence-transformers/quora-distilbert-base
-    - sentence-transformers/quora-distilbert-multilingual
-    - sentence-transformers/sentence-t5-base
-    - sentence-transformers/sentence-t5-large
-    - sentence-transformers/sentence-t5-xxl
-    - sentence-transformers/sentence-t5-xl
-    - sentence-transformers/stsb-distilroberta-base-v2
-    - sentence-transformers/stsb-mpnet-base-v2
-    - sentence-transformers/stsb-roberta-base-v2
-    - sentence-transformers/stsb-xlm-r-multilingual
-    - sentence-transformers/xlm-r-distilroberta-base-paraphrase-v1
-    - sentence-transformers/clip-ViT-L-14
-    - sentence-transformers/clip-ViT-B-16
-    - sentence-transformers/use-cmlm-multilingual
-    - sentence-transformers/all-MiniLM-L12-v1
-    ```
-
-!!! info
-    You can also load many other model architectures from the library. For example models from sources such as BAAI, nomic, salesforce research, etc.
-    See this HF hub page for all [supported models](https://huggingface.co/models?library=sentence-transformers).
-
-!!! note "BAAI Embeddings example"
-    Here is an example that uses BAAI embedding model from the HuggingFace Hub [supported models](https://huggingface.co/models?library=sentence-transformers)
+!!! example "Example usage"
     ```python
-    import lancedb
-    from lancedb.pydantic import LanceModel, Vector
-    from lancedb.embeddings import get_registry
-
-    db = lancedb.connect("/tmp/db")
-    model = get_registry().get("sentence-transformers").create(name="BAAI/bge-small-en-v1.5", device="cpu")
-
-    class Words(LanceModel):
-        text: str = model.SourceField()
-        vector: Vector(model.ndims()) = model.VectorField()
-
-    table = db.create_table("words", schema=Words)
-    table.add(
-        [
-            {"text": "hello world"},
-            {"text": "goodbye world"}
-        ]
-    )
-
-    query = "greetings"
-    actual = table.search(query).limit(1).to_pydantic(Words)[0]
-    print(actual.text)
-    ```
-Visit sentence-transformers [HuggingFace HUB](https://huggingface.co/sentence-transformers) page for more information on the available models.
-
-
-### Huggingface embedding models
-We offer support for all huggingface models (which can be loaded via [transformers](https://huggingface.co/docs/transformers/en/index) library). The default model is `colbert-ir/colbertv2.0` which also has its own special callout - `registry.get("colbert")` 
-
-Example usage - 
-```python
-import lancedb
-import pandas as pd
-
-from lancedb.embeddings import get_registry
-from lancedb.pydantic import LanceModel, Vector
-
-model = get_registry().get("huggingface").create(name='facebook/bart-base')
-
-class Words(LanceModel):
-    text: str = model.SourceField()
-    vector: Vector(model.ndims()) = model.VectorField()
-
-df = pd.DataFrame({"text": ["hi hello sayonara", "goodbye world"]})
-table = db.create_table("greets", schema=Words)
-table.add(df)
-query = "old greeting"
-actual = table.search(query).limit(1).to_pydantic(Words)[0]
-print(actual.text)
-```
-
-
-### Ollama embeddings
-Generate embeddings via the [ollama](https://github.com/ollama/ollama-python) python library. More details:
-
-- [Ollama docs on embeddings](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embeddings)
-- [Ollama blog on embeddings](https://ollama.com/blog/embedding-models)
-
-| Parameter              | Type                       | Default Value            | Description                                                                                                                                    |
-|------------------------|----------------------------|--------------------------|------------------------------------------------------------------------------------------------------------------------------------------------|
-| `name`                 | `str`                      | `nomic-embed-text`       | The name of the model.                                                                                                                         |
-| `host`                 | `str`                      | `http://localhost:11434` | The Ollama host to connect to.                                                                                                                 |
-| `options`              | `ollama.Options` or `dict` | `None`                   | Additional model parameters listed in the documentation for the Modelfile such as `temperature`. |
-| `keep_alive`           | `float` or `str`           | `"5m"`                   | Controls how long the model will stay loaded into memory following the request.                                                                |
-| `ollama_client_kwargs` | `dict`                     | `{}`                     | kwargs that can be past to the `ollama.Client`.                                                                                                |
-
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-
-db = lancedb.connect("/tmp/db")
-func = get_registry().get("ollama").create(name="nomic-embed-text")
-
-class Words(LanceModel):
-    text: str = func.SourceField()
-    vector: Vector(func.ndims()) = func.VectorField()
-
-table = db.create_table("words", schema=Words, mode="overwrite")
-table.add([
-    {"text": "hello world"},
-    {"text": "goodbye world"}
-])
-
-query = "greetings"
-actual = table.search(query).limit(1).to_pydantic(Words)[0]
-print(actual.text)
-```
-
-
-### OpenAI embeddings
-LanceDB registers the OpenAI embeddings function in the registry by default, as `openai`. Below are the parameters that you can customize when creating the instances:
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"text-embedding-ada-002"` | The name of the model. |
-| `dim` | `int` |  Model default   | For OpenAI's newer text-embedding-3 model, we can specify a dimensionality that is smaller than the 1536 size. This feature supports it |
-
-
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-
-db = lancedb.connect("/tmp/db")
-func = get_registry().get("openai").create(name="text-embedding-ada-002")
-
-class Words(LanceModel):
-    text: str = func.SourceField()
-    vector: Vector(func.ndims()) = func.VectorField()
-
-table = db.create_table("words", schema=Words, mode="overwrite")
-table.add(
-    [
-        {"text": "hello world"},
-        {"text": "goodbye world"}
-    ]
-    )
-
-query = "greetings"
-actual = table.search(query).limit(1).to_pydantic(Words)[0]
-print(actual.text)
-```
-
-### Instructor Embeddings
-[Instructor](https://instructor-embedding.github.io/) is an instruction-finetuned text embedding model that can generate text embeddings tailored to any task (e.g. classification, retrieval, clustering, text evaluation, etc.) and domains (e.g. science, finance, etc.) by simply providing the task instruction, without any finetuning.
-
-If you want to calculate customized embeddings for specific sentences, you can follow the unified template to write instructions.
-
-!!! info
-    Represent the `domain` `text_type` for `task_objective`:
-
-    * `domain` is optional, and it specifies the domain of the text, e.g. science, finance, medicine, etc.
-    * `text_type` is required, and it specifies the encoding unit, e.g. sentence, document, paragraph, etc.
-    * `task_objective` is optional, and it specifies the objective of embedding, e.g. retrieve a document, classify the sentence, etc.
-
-More information about the model can be found at the [source URL](https://github.com/xlang-ai/instructor-embedding).
-
-| Argument | Type | Default | Description |
-|---|---|---|---|
-| `name` | `str` | "hkunlp/instructor-base" | The name of the model to use |
-| `batch_size` | `int` | `32` | The batch size to use when generating embeddings |
-| `device` | `str` | `"cpu"` | The device to use when generating embeddings |
-| `show_progress_bar` | `bool` | `True` | Whether to show a progress bar when generating embeddings |
-| `normalize_embeddings` | `bool` | `True` | Whether to normalize the embeddings |
-| `quantize` | `bool` | `False` | Whether to quantize the model |
-| `source_instruction` | `str` | `"represent the docuement for retreival"` | The instruction for the source column |
-| `query_instruction` | `str` | `"represent the document for retreiving the most similar documents"` | The instruction for the query |
-
-
-
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry, InstuctorEmbeddingFunction
-
-instructor = get_registry().get("instructor").create(
-                            source_instruction="represent the docuement for retreival",
-                            query_instruction="represent the document for retreiving the most similar documents"
-                            )
-
-class Schema(LanceModel):
-    vector: Vector(instructor.ndims()) = instructor.VectorField()
-    text: str = instructor.SourceField()
-
-db = lancedb.connect("~/.lancedb")
-tbl = db.create_table("test", schema=Schema, mode="overwrite")
-
-texts = [{"text": "Capitalism has been dominant in the Western world since the end of feudalism, but most feel[who?] that..."},
-        {"text": "The disparate impact theory is especially controversial under the Fair Housing Act because the Act..."},
-        {"text": "Disparate impact in United States labor law refers to practices in employment, housing, and other areas that.."}]
-
-tbl.add(texts)
-```
-
-### Gemini Embeddings
-With Google's Gemini, you can represent text (words, sentences, and blocks of text) in a vectorized form, making it easier to compare and contrast embeddings. For example, two texts that share a similar subject matter or sentiment should have similar embeddings, which can be identified through mathematical comparison techniques such as cosine similarity. For more on how and why you should use embeddings, refer to the Embeddings guide.
-The Gemini Embedding Model API supports various task types:
-
-| Task Type               | Description                                                                                                                                                |
-|-------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| "`retrieval_query`"     | Specifies the given text is a query in a search/retrieval setting.                                                                                         |
-| "`retrieval_document`"  | Specifies the given text is a document in a search/retrieval setting. Using this task type requires a title but is automatically proided by Embeddings API |
-| "`semantic_similarity`" | Specifies the given text will be used for Semantic Textual Similarity (STS).                                                                               |
-| "`classification`"      | Specifies that the embeddings will be used for classification.                                                                                             |
-| "`clusering`"           | Specifies that the embeddings will be used for clustering.                                                                                                 |
-
-
-Usage Example:
-
-```python
-import lancedb
-import pandas as pd
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-
-
-model = get_registry().get("gemini-text").create()
-
-class TextModel(LanceModel):
-    text: str = model.SourceField()
-    vector: Vector(model.ndims()) = model.VectorField()
-
-df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
-db = lancedb.connect("~/.lancedb")
-tbl = db.create_table("test", schema=TextModel, mode="overwrite")
-
-tbl.add(df)
-rs = tbl.search("hello").limit(1).to_pandas()
-```
-
-### Cohere Embeddings
-Using cohere API requires cohere package, which can be installed using `pip install cohere`. Cohere embeddings are used to generate embeddings for text data. The embeddings can be used for various tasks like semantic search, clustering, and classification.
-You also need to set the `COHERE_API_KEY` environment variable to use the Cohere API.
-
-Supported models are:
-```
-    * embed-english-v3.0
-    * embed-multilingual-v3.0
-    * embed-english-light-v3.0
-    * embed-multilingual-light-v3.0
-    * embed-english-v2.0
-    * embed-english-light-v2.0
-    * embed-multilingual-v2.0
-```
-
-Supported parameters (to be passed in `create` method) are:
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"embed-english-v2.0"` | The model ID of the cohere model to use. Supported base models for Text Embeddings: embed-english-v3.0, embed-multilingual-v3.0, embed-english-light-v3.0, embed-multilingual-light-v3.0, embed-english-v2.0, embed-english-light-v2.0, embed-multilingual-v2.0 |
-| `source_input_type` | `str` | `"search_document"` | The type of input data to be used for the source column. |
-| `query_input_type` | `str` | `"search_query"` | The type of input data to be used for the query. |
-
-Cohere supports following input types:
-| Input Type               | Description                          |
-|-------------------------|---------------------------------------|
-| "`search_document`"     | Used for embeddings stored in a vector|
-|                         | database for search use-cases.        |
-| "`search_query`"        | Used for embeddings of search queries |
-|                         | run against a vector DB               |
-| "`semantic_similarity`" | Specifies the given text will be used |
-|                         | for Semantic Textual Similarity (STS) |
-| "`classification`"      | Used for embeddings passed through a  |
-|                         | text classifier.                      |
-| "`clustering`"          | Used for the embeddings run through a |
-|                         | clustering algorithm                  |
-
-Usage Example:
-    
-    ```python
-        import lancedb
-        from lancedb.pydantic import LanceModel, Vector
-        from lancedb.embeddings import EmbeddingFunctionRegistry
-
-        cohere = EmbeddingFunctionRegistry
-            .get_instance()
-            .get("cohere")
-            .create(name="embed-multilingual-v2.0")
-
-        class TextModel(LanceModel):
-            text: str = cohere.SourceField()
-            vector: Vector(cohere.ndims()) =  cohere.VectorField()
-
-        data = [ { "text": "hello world" },
-                { "text": "goodbye world" }]
-
-        db = lancedb.connect("~/.lancedb")
-        tbl = db.create_table("test", schema=TextModel, mode="overwrite")
-
-        tbl.add(data)
+    model = get_registry()
+              .get("openai")
+              .create(name="text-embedding-ada-002")
     ```
 
-### Jina Embeddings
-Jina embeddings are used to generate embeddings for text and image data.
-You also need to set the `JINA_API_KEY` environment variable to use the Jina API.
-
-You can find a list of supported models under [https://jina.ai/embeddings/](https://jina.ai/embeddings/)
-
-Supported parameters (to be passed in `create` method) are:
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"jina-clip-v1"` | The model ID of the jina model to use |
-
-Usage Example:
-
+Now let's understand the above syntax: 
 ```python
-    import os
-    import lancedb
-    from lancedb.pydantic import LanceModel, Vector
-    from lancedb.embeddings import EmbeddingFunctionRegistry
-
-    os.environ['JINA_API_KEY'] = 'jina_*'
-
-    jina_embed = EmbeddingFunctionRegistry.get_instance().get("jina").create(name="jina-embeddings-v2-base-en")
-
-
-    class TextModel(LanceModel):
-        text: str = jina_embed.SourceField()
-        vector: Vector(jina_embed.ndims()) = jina_embed.VectorField()
-
-
-    data = [{"text": "hello world"},
-            {"text": "goodbye world"}]
-
-    db = lancedb.connect("~/.lancedb-2")
-    tbl = db.create_table("test", schema=TextModel, mode="overwrite")
-
-    tbl.add(data)
+model = get_registry().get("model_id").create(...params)
 ```
+**This👆 line effectively creates a configured instance of an `embedding function` with `model` of choice that is ready for use.**
 
-### AWS Bedrock Text Embedding Functions
-AWS Bedrock supports multiple base models for generating text embeddings. You need to setup the AWS credentials to use this embedding function.
-You can do so by using `awscli` and also add your session_token:
-```shell
-aws configure
-aws configure set aws_session_token "<your_session_token>"
-```
-to ensure that the credentials are set up correctly, you can run the following command:
-```shell
-aws sts get-caller-identity
-```
+- `get_registry()` :  This function call returns an instance of a `EmbeddingFunctionRegistry` object. This registry manages the registration and retrieval of embedding functions.
 
-Supported Embedding modelIDs are:
-* `amazon.titan-embed-text-v1`
-* `cohere.embed-english-v3`
-* `cohere.embed-multilingual-v3`
+- `.get("model_id")` : This method call on the registry object and retrieves the **embedding models functions** associated with the `"model_id"` (1) .
+    { .annotate }
 
-Supported parameters (to be passed in `create` method) are:
+    1.  Hover over the names in table below to find out the `model_id` of different embedding functions.
 
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| **name** | str | "amazon.titan-embed-text-v1" | The model ID of the bedrock model to use. Supported base models for Text Embeddings: amazon.titan-embed-text-v1, cohere.embed-english-v3, cohere.embed-multilingual-v3 |
-| **region** | str | "us-east-1" | Optional name of the AWS Region in which the service should be called (e.g., "us-east-1"). |
-| **profile_name** | str | None | Optional name of the AWS profile to use for calling the Bedrock service. If not specified, the default profile will be used. |
-| **assumed_role** | str | None | Optional ARN of an AWS IAM role to assume for calling the Bedrock service. If not specified, the current active credentials will be used. |
-| **role_session_name** | str | "lancedb-embeddings" | Optional name of the AWS IAM role session to use for calling the Bedrock service. If not specified, a "lancedb-embeddings" name will be used. |
-| **runtime** | bool | True | Optional choice of getting different client to perform operations with the Amazon Bedrock service. |
-| **max_retries** | int | 7 | Optional number of retries to perform when a request fails. |
+- `.create(...params)` : This method call is on the object returned by the `get` method. It instantiates an embedding model function using the **specified parameters**. 
 
-Usage Example:
+??? question "What parameters does the `.create(...params)` method accepts?"
+    **Checkout the documentation of specific embedding models (links in the table below👇) to know what parameters it takes**.
 
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
+!!! tip "Moving on"
+    Now that we know how to get the **desired embedding model** and use it in our code, let's explore the comprehensive **list** of embedding models **supported by LanceDB**, in the tables below.
 
-model = get_registry().get("bedrock-text").create()
+## Text Embedding Functions 📝 
+These functions are registered by default to handle text embeddings.
 
-class TextModel(LanceModel):
-    text: str = model.SourceField()
-    vector: Vector(model.ndims()) = model.VectorField()
+- 🔄 **Embedding functions** have an inbuilt rate limit handler wrapper for source and query embedding function calls that retry with **exponential backoff**. 
 
-df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
-db = lancedb.connect("tmp_path")
-tbl = db.create_table("test", schema=TextModel, mode="overwrite")
+- 🌕 Each `EmbeddingFunction` implementation automatically takes `max_retries` as an argument which has the default value of 7. 
 
-tbl.add(df)
-rs = tbl.search("hello").limit(1).to_pandas()
-```
+🌟 **Available Text Embeddings**
 
-## Multi-modal embedding functions
-Multi-modal embedding functions allow you to query your table using both images and text.
-
-### OpenClip embeddings
-We support CLIP model embeddings using the open source alternative, [open-clip](https://github.com/mlfoundations/open_clip) which supports various customizations. It is registered as `open-clip` and supports the following customizations:
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"ViT-B-32"` | The name of the model. |
-| `pretrained` | `str` | `"laion2b_s34b_b79k"` | The name of the pretrained model to load. |
-| `device` | `str` | `"cpu"` | The device to run the model on. Can be `"cpu"` or `"gpu"`. |
-| `batch_size` | `int` | `64` | The number of images to process in a batch. |
-| `normalize` | `bool` | `True` | Whether to normalize the input images before feeding them to the model. |
-
-This embedding function supports ingesting images as both bytes and urls. You can query them using both test and other images.
-
-!!! info
-    LanceDB supports ingesting images directly from accessible links.
-
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-
-db = lancedb.connect(tmp_path)
-func = get_registry.get("open-clip").create()
-
-class Images(LanceModel):
-    label: str
-    image_uri: str = func.SourceField() # image uri as the source
-    image_bytes: bytes = func.SourceField() # image bytes as the source
-    vector: Vector(func.ndims()) = func.VectorField() # vector column 
-    vec_from_bytes: Vector(func.ndims()) = func.VectorField() # Another vector column 
-
-table = db.create_table("images", schema=Images)
-labels = ["cat", "cat", "dog", "dog", "horse", "horse"]
-uris = [
-    "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
-    "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
-    "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
-    "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
-    "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
-    "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
-]
-# get each uri as bytes
-image_bytes = [requests.get(uri).content for uri in uris]
-table.add(
-    pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": image_bytes})
-)
-```
-Now we can search using text from both the default vector column and the custom vector column
-```python
-
-# text search
-actual = table.search("man's best friend").limit(1).to_pydantic(Images)[0]
-print(actual.label) # prints "dog"
-
-frombytes = (
-    table.search("man's best friend", vector_column_name="vec_from_bytes")
-    .limit(1)
-    .to_pydantic(Images)[0]
-)
-print(frombytes.label)
-
-```
-
-Because we're using a multi-modal embedding function, we can also search using images
-
-```python
-# image search
-query_image_uri = "http://farm1.staticflickr.com/200/467715466_ed4a31801f_z.jpg"
-image_bytes = requests.get(query_image_uri).content
-query_image = Image.open(io.BytesIO(image_bytes))
-actual = table.search(query_image).limit(1).to_pydantic(Images)[0]
-print(actual.label == "dog")
-
-# image search using a custom vector column
-other = (
-    table.search(query_image, vector_column_name="vec_from_bytes")
-    .limit(1)
-    .to_pydantic(Images)[0]
-)
-print(actual.label)
-
-```
-
-### Imagebind embeddings
-We have support for [imagebind](https://github.com/facebookresearch/ImageBind) model embeddings. You can download our version of the packaged model via - `pip install imagebind-packaged==0.1.2`.
-
-This function is registered as `imagebind` and supports Audio, Video and Text modalities(extending to Thermal,Depth,IMU data):
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"imagebind_huge"` | Name of the model. |
-| `device` | `str` | `"cpu"` | The device to run the model on. Can be `"cpu"` or `"gpu"`. |
-| `normalize` | `bool` | `False` | set to `True` to normalize your inputs before model ingestion. |
-
-Below is an example demonstrating how the API works:
-
-```python
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-
-db = lancedb.connect(tmp_path)
-func = get_registry.get("imagebind").create()
-
-class ImageBindModel(LanceModel):
-    text: str
-    image_uri: str = func.SourceField()
-    audio_path: str
-    vector: Vector(func.ndims()) = func.VectorField()
-
-# add locally accessible image paths
-text_list=["A dog.", "A car", "A bird"]
-image_paths=[".assets/dog_image.jpg", ".assets/car_image.jpg", ".assets/bird_image.jpg"]
-audio_paths=[".assets/dog_audio.wav", ".assets/car_audio.wav", ".assets/bird_audio.wav"]
-
-# Load data
-inputs = [
-    {"text": a, "audio_path": b, "image_uri": c}
-    for a, b, c in zip(text_list, audio_paths, image_paths)
-]
-
-#create table and add data
-table = db.create_table("img_bind", schema=ImageBindModel)
-table.add(inputs)
-```
-
-Now, we can search using any modality:
-
-#### image search
-```python
-query_image = "./assets/dog_image2.jpg" #download an image and enter that path here
-actual = table.search(query_image).limit(1).to_pydantic(ImageBindModel)[0]
-print(actual.text == "dog")
-```
-#### audio search
-
-```python
-query_audio = "./assets/car_audio2.wav" #download an audio clip and enter path here
-actual = table.search(query_audio).limit(1).to_pydantic(ImageBindModel)[0]
-print(actual.text == "car")
-```
-#### Text search
-You can add any input query and fetch the result as follows:
-```python
-query = "an animal which flies and tweets" 
-actual = table.search(query).limit(1).to_pydantic(ImageBindModel)[0]
-print(actual.text == "bird")
-```
-
-If you have any questions about the embeddings API, supported models, or see a relevant model missing, please raise an issue [on GitHub](https://github.com/lancedb/lancedb/issues).
-
-### Jina Embeddings
-Jina embeddings can also be used to embed both text and image data, only some of the models support image data and you can check the list
-under [https://jina.ai/embeddings/](https://jina.ai/embeddings/)
-
-Supported parameters (to be passed in `create` method) are:
-
-| Parameter | Type | Default Value | Description |
-|---|---|---|---|
-| `name` | `str` | `"jina-clip-v1"` | The model ID of the jina model to use |
-
-Usage Example:
-
-```python
-    import os
-    import requests
-    import lancedb
-    from lancedb.pydantic import LanceModel, Vector
-    from lancedb.embeddings import get_registry
-    import pandas as pd
-
-    os.environ['JINA_API_KEY'] = 'jina_*'
-
-    db = lancedb.connect("~/.lancedb")
-    func = get_registry().get("jina").create()
+| **Embedding** :material-information-outline:{ title="Hover over the name to find out the model_id" } | **Description** | **Documentation** |
+|-----------|-------------|---------------|
+| [**Sentence Transformers**](available_embedding_models/text_embedding_functions/sentence_transformers.md "sentence-transformers")  | 🧠 **SentenceTransformers** is a Python framework for state-of-the-art sentence, text, and image embeddings. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/sbert_2.png" alt="Sentence Transformers Icon" width="90" height="35">](available_embedding_models/text_embedding_functions/sentence_transformers.md)|                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Huggingface Models**](available_embedding_models/text_embedding_functions/huggingface_embedding.md "huggingface") |🤗 We offer support for all **Huggingface** models. The default model is `colbert-ir/colbertv2.0`. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/hugging_face.png" alt="Huggingface Icon" width="130" height="35">](available_embedding_models/text_embedding_functions/huggingface_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Ollama Embeddings**](available_embedding_models/text_embedding_functions/ollama_embedding.md "ollama") | 🔍 Generate embeddings via the **Ollama** python library. Ollama supports embedding models, making it possible to build RAG apps. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/Ollama.png" alt="Ollama Icon" width="110" height="35">](available_embedding_models/text_embedding_functions/ollama_embedding.md)|                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**OpenAI Embeddings**](available_embedding_models/text_embedding_functions/openai_embedding.md "openai")| 🔑 **OpenAI’s** text embeddings measure the relatedness of text strings. **LanceDB** supports state-of-the-art embeddings from OpenAI. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/openai.png" alt="OpenAI Icon" width="100" height="35">](available_embedding_models/text_embedding_functions/openai_embedding.md)|                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Instructor Embeddings**](available_embedding_models/text_embedding_functions/instructor_embedding.md "instructor") | 📚 **Instructor**: An instruction-finetuned text embedding model that can generate text embeddings tailored to any task and domains by simply providing the task instruction, without any finetuning. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/instructor_embedding.png" alt="Instructor Embedding Icon" width="140" height="35">](available_embedding_models/text_embedding_functions/instructor_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Gemini Embeddings**](available_embedding_models/text_embedding_functions/gemini_embedding.md "gemini-text") | 🌌 Google’s Gemini API generates state-of-the-art embeddings for words, phrases, and sentences. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/gemini.png" alt="Gemini Icon" width="95" height="35">](available_embedding_models/text_embedding_functions/gemini_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Cohere Embeddings**](available_embedding_models/text_embedding_functions/cohere_embedding.md "cohere") | 💬 This will help you get started with **Cohere** embedding models using LanceDB. Using cohere API requires cohere package. Install it via `pip`. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/cohere.png" alt="Cohere Icon" width="140" height="35">](available_embedding_models/text_embedding_functions/cohere_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**Jina Embeddings**](available_embedding_models/text_embedding_functions/jina_embedding.md "jina") | 🔗 World-class embedding models to improve your search and RAG systems. You will need **jina api key**. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/jina.png" alt="Jina Icon" width="90" height="35">](available_embedding_models/text_embedding_functions/jina_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [ **AWS Bedrock Functions**](available_embedding_models/text_embedding_functions/aws_bedrock_embedding.md "bedrock-text") | ☁️ AWS Bedrock supports multiple base models for generating text embeddings. You need to setup the AWS credentials to use this embedding function. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/aws_bedrock.png" alt="AWS Bedrock Icon" width="120" height="35">](available_embedding_models/text_embedding_functions/aws_bedrock_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
+| [**IBM Watsonx.ai**](available_embedding_models/text_embedding_functions/ibm_watsonx_ai_embedding.md "watsonx") | 💡 Generate text embeddings using IBM's watsonx.ai platform. **Note**: watsonx.ai library is an optional dependency. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/watsonx.png" alt="Watsonx Icon" width="140" height="35">](available_embedding_models/text_embedding_functions/ibm_watsonx_ai_embedding.md) |
+| [**VoyageAI Embeddings**](available_embedding_models/text_embedding_functions/voyageai_embedding.md "voyageai") | 🌕 Voyage AI provides cutting-edge embedding and rerankers. This will help you get started with **VoyageAI** embedding models using LanceDB. Using voyageai API requires voyageai package. Install it via `pip`. | [<img src="https://www.voyageai.com/logo.svg" alt="VoyageAI Icon" width="140" height="35">](available_embedding_models/text_embedding_functions/voyageai_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
 
 
-    class Images(LanceModel):
-        label: str
-        image_uri: str = func.SourceField()  # image uri as the source
-        image_bytes: bytes = func.SourceField()  # image bytes as the source
-        vector: Vector(func.ndims()) = func.VectorField()  # vector column
-        vec_from_bytes: Vector(func.ndims()) = func.VectorField()  # Another vector column
+
+[st-key]: "sentence-transformers"
+[hf-key]: "huggingface"
+[ollama-key]: "ollama"
+[openai-key]: "openai"
+[instructor-key]: "instructor"
+[gemini-key]: "gemini-text"
+[cohere-key]: "cohere"
+[jina-key]: "jina"
+[aws-key]: "bedrock-text"
+[watsonx-key]: "watsonx"
+[voyageai-key]: "voyageai"
 
 
-    table = db.create_table("images", schema=Images)
-    labels = ["cat", "cat", "dog", "dog", "horse", "horse"]
-    uris = [
-        "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
-        "http://farm1.staticflickr.com/134/332220238_da527d8140_z.jpg",
-        "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
-        "http://farm5.staticflickr.com/4092/5017326486_1f46057f5f_z.jpg",
-        "http://farm9.staticflickr.com/8216/8434969557_d37882c42d_z.jpg",
-        "http://farm6.staticflickr.com/5142/5835678453_4f3a4edb45_z.jpg",
-    ]
-    # get each uri as bytes
-    image_bytes = [requests.get(uri).content for uri in uris]
-    table.add(
-      pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": image_bytes})
-    )
-```
+## Multi-modal Embedding Functions🖼️ 
+
+Multi-modal embedding functions allow you to query your table using both images and text. 💬🖼️
+
+🌐 **Available Multi-modal Embeddings**
+
+| Embedding :material-information-outline:{ title="Hover over the name to find out the model_id" }  | Description | Documentation  |
+|-----------|-------------|---------------|
+| [**OpenClip Embeddings**](available_embedding_models/multimodal_embedding_functions/openclip_embedding.md "open-clip") | 🎨 We support CLIP model embeddings using the open source alternative, **open-clip** which supports various customizations. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/openclip_github.png" alt="openclip Icon" width="150" height="35">](available_embedding_models/multimodal_embedding_functions/openclip_embedding.md) |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    
+| [**Imagebind Embeddings**](available_embedding_models/multimodal_embedding_functions/imagebind_embedding.md "imageind") | 🌌  We have support for **imagebind model embeddings**. You can download our version of the packaged model via - `pip install imagebind-packaged==0.1.2`. | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/imagebind_meta.png" alt="imagebind Icon" width="150" height="35">](available_embedding_models/multimodal_embedding_functions/imagebind_embedding.md)|                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    
+| [**Jina Multi-modal Embeddings**](available_embedding_models/multimodal_embedding_functions/jina_multimodal_embedding.md "jina") | 🔗 **Jina embeddings** can also be used to embed both **text** and **image** data, only some of the models support image data and you can check the detailed documentation. 👉 | [<img src="https://raw.githubusercontent.com/lancedb/assets/main/docs/assets/logos/jina.png" alt="jina Icon" width="90" height="35">](available_embedding_models/multimodal_embedding_functions/jina_multimodal_embedding.md) |
+
+!!! note
+    If you'd like to request support for additional **embedding functions**, please feel free to open an issue on our LanceDB [GitHub issue page](https://github.com/lancedb/lancedb/issues).

docs/src/embeddings/embedding_functions.md

@@ -2,8 +2,8 @@ Representing multi-modal data as vector embeddings is becoming a standard practi
 
 For this purpose, LanceDB introduces an **embedding functions API**, that allow you simply set up once, during the configuration stage of your project. After this, the table remembers it, effectively making the embedding functions *disappear in the background* so you don't have to worry about manually passing callables, and instead, simply focus on the rest of your data engineering pipeline.
 
-!!! Note "LanceDB cloud doesn't support embedding functions yet"
-    LanceDB Cloud does not support embedding functions yet. You need to generate embeddings before ingesting into the table or querying.
+!!! Note "Embedding functions on LanceDB cloud"
+    When using embedding functions with LanceDB cloud, the embeddings will be generated on the source device and sent to the cloud. This means that the source device must have the necessary resources to generate the embeddings.
 
 !!! warning
     Using the embedding function registry means that you don't have to explicitly generate the embeddings yourself.
@@ -94,8 +94,8 @@ the embeddings at all:
     === "@lancedb/lancedb"
 
         ```ts
-        --8<-- "nodejs/examples/embedding.ts:imports"
-        --8<-- "nodejs/examples/embedding.ts:embedding_function"
+        --8<-- "nodejs/examples/embedding.test.ts:imports"
+        --8<-- "nodejs/examples/embedding.test.ts:embedding_function"
         ```
 
     === "vectordb (deprecated)"
@@ -150,7 +150,7 @@ need to worry about it when you query the table:
             .toArray()
         ```
 
-    === "vectordb (deprecated)
+    === "vectordb (deprecated)"
 
         ```ts
         const results = await table

docs/src/embeddings/index.md

@@ -51,8 +51,8 @@ LanceDB registers the OpenAI embeddings function in the registry as `openai`. Yo
 === "TypeScript"
 
     ```typescript
-    --8<--- "nodejs/examples/embedding.ts:imports"
-    --8<--- "nodejs/examples/embedding.ts:openai_embeddings"
+    --8<--- "nodejs/examples/embedding.test.ts:imports"
+    --8<--- "nodejs/examples/embedding.test.ts:openai_embeddings"
     ```
 
 === "Rust"
@@ -99,34 +99,32 @@ LanceDB registers the Sentence Transformers embeddings function in the registry
 
     Coming Soon!
 
-### Jina Embeddings
-
-LanceDB registers the JinaAI embeddings function in the registry as `jina`. You can pass any supported model name to the `create`. By default it uses `"jina-clip-v1"`.
-`jina-clip-v1` can handle both text and images and other models only support `text`.
-
-You need to pass `JINA_API_KEY` in the environment variable or pass it as `api_key` to `create` method.
+### Embedding function with LanceDB cloud
+Embedding functions are now supported on LanceDB cloud. The embeddings will be generated on the source device and sent to the cloud. This means that the source device must have the necessary resources to generate the embeddings. Here's an example using the OpenAI embedding function:
 
 ```python
 import os
 import lancedb
 from lancedb.pydantic import LanceModel, Vector
 from lancedb.embeddings import get_registry
-os.environ['JINA_API_KEY'] = "jina_*"
+os.environ['OPENAI_API_KEY'] = "..."
 
-db = lancedb.connect("/tmp/db")
-func = get_registry().get("jina").create(name="jina-clip-v1")
+db = lancedb.connect(
+  uri="db://....",
+  api_key="sk_...",
+  region="us-east-1"
+)
+func = get_registry().get("openai").create()
 
 class Words(LanceModel):
     text: str = func.SourceField()
     vector: Vector(func.ndims()) = func.VectorField()
 
-table = db.create_table("words", schema=Words, mode="overwrite")
-table.add(
-    [
-        {"text": "hello world"},
-        {"text": "goodbye world"}
-    ]
-    )
+table = db.create_table("words", schema=Words)
+table.add([
+    {"text": "hello world"},
+    {"text": "goodbye world"}
+])
 
 query = "greetings"
 actual = table.search(query).limit(1).to_pydantic(Words)[0]

docs/src/embeddings/understanding_embeddings.md

@@ -0,0 +1,133 @@
+# Understand Embeddings
+
+The term **dimension** is a synonym for the number of elements in a feature vector. Each feature can be thought of as a different axis in a geometric space. 
+
+High-dimensional data means there are many features(or attributes) in the data.
+
+!!! example
+     1. An image is a data point and it might have thousands of dimensions because each pixel could be considered as a feature. 
+
+     2. Text data, when represented by each word or character, can also lead to high dimensions, especially when considering all possible words in a language.
+
+Embedding captures **meaning and relationships** within data by mapping high-dimensional data into a lower-dimensional space. It captures it by placing inputs that are more **similar in meaning** closer together in the **embedding space**. 
+
+## What are Vector Embeddings?
+
+Vector embeddings is a way to convert complex data, like text, images, or audio into numerical coordinates (called vectors) that can be plotted in an n-dimensional space(embedding space). 
+
+The closer these data points are related in the real world, the closer their corresponding numerical coordinates (vectors) will be to each other in the embedding space. This proximity in the embedding space reflects their semantic similarities, allowing machines to intuitively understand and process the data in a way that mirrors human perception of relationships and meaning.
+
+In a way, it captures the most important aspects of the data while ignoring the less important ones. As a result, tasks like searching for related content or identifying patterns become more efficient and accurate, as the embeddings make it possible to quantify how **closely related** different **data points** are and **reduce** the **computational complexity**.
+
+??? question "Are vectors and embeddings the same thing?"
+
+    When we say “vectors” we mean - **list of numbers** that **represents the data**. 
+    When we say “embeddings” we mean - **list of numbers** that **capture important details and relationships**.
+
+    Although the terms are often used interchangeably, “embeddings” highlight how the data is represented with meaning and structure, while “vector” simply refers to the numerical form of that representation.
+
+## Embedding vs Indexing
+
+We already saw that creating **embeddings** on data is a method of creating **vectors** for a **n-dimensional embedding space** that captures the meaning and relationships inherent in the data.
+
+Once we have these **vectors**, indexing comes into play. Indexing is a method of organizing these vector embeddings, that allows us to quickly and efficiently locate and retrieve them from the entire dataset of vector embeddings.
+
+## What types of data/objects can be embedded?
+
+The following are common types of data that can be embedded:
+
+1. **Text**: Text data includes sentences, paragraphs, documents, or any written content.
+2. **Images**:  Image data encompasses photographs, illustrations, or any visual content.
+3. **Audio**: Audio data includes sounds, music, speech, or any auditory content.
+4. **Video**:  Video data consists of moving images and sound, which can convey complex information.
+
+Large datasets of multi-modal data (text, audio, images, etc.) can be converted into embeddings with the appropriate model.
+
+!!! tip "LanceDB vs Other traditional Vector DBs"
+    While many vector databases primarily focus on the storage and retrieval of vector embeddings, **LanceDB** uses **Lance file format** (operates on a disk-based architecture), which allows for the storage and management of not just embeddings but also **raw file data (bytes)**. This capability means that users can integrate various types of data, including images and text, alongside their vector embeddings in a unified system.
+
+    With the ability to store both vectors and associated file data, LanceDB enhances the querying process. Users can perform semantic searches that not only retrieve similar embeddings but also access related files and metadata, thus streamlining the workflow.
+
+## How does embedding works?
+
+As mentioned, after creating embedding, each data point is represented as a vector in a n-dimensional space (embedding space). The dimensionality of this space can vary depending on the complexity of the data and the specific embedding technique used.
+
+Points that are close to each other in vector space are considered similar (or appear in similar contexts), and points that are far away are considered dissimilar. To quantify this closeness, we use distance as a metric which can be measured in the  following way - 
+
+1. **Euclidean Distance (l2)**: It calculates the straight-line distance between two points (vectors) in a multidimensional space.
+2. **Cosine Similarity**: It measures the cosine of the angle between two vectors, providing a normalized measure of similarity based on their direction.
+3. **Dot product**: It is calculated as the sum of the products of their corresponding components. To measure relatedness it considers both the magnitude and direction of the vectors.
+
+## How do you create and store vector embeddings for your data?
+
+1. **Creating embeddings**: Choose an embedding model, it can be a pre-trained model (open-source or commercial) or you can train a custom embedding model for your scenario. Then feed your preprocessed data into the chosen model to obtain embeddings.
+
+??? question "Popular choices for embedding models"
+    For text data, popular choices are OpenAI’s text-embedding models, Google Gemini text-embedding models, Cohere’s Embed models, and SentenceTransformers, etc.
+
+    For image data, popular choices are CLIP (Contrastive Language–Image Pretraining), Imagebind embeddings by meta (supports audio, video, and image), and Jina multi-modal embeddings, etc.
+
+2. **Storing vector embeddings**: This effectively requires **specialized databases** that can handle the complexity of vector data, as traditional databases often struggle with this task. Vector databases are designed specifically for storing and querying vector embeddings. They optimize for efficient nearest-neighbor searches and provide built-in indexing mechanisms.
+
+!!! tip "Why LanceDB"
+    LanceDB **automates** the entire process of creating and storing embeddings for your data. LanceDB allows you to define and use **embedding functions**, which can be **pre-trained models** or **custom models**. 
+    
+    This enables you to **generate** embeddings tailored to the nature of your data (e.g., text, images) and **store** both the **original data** and **embeddings** in a **structured schema** thus providing efficient querying capabilities for similarity searches.
+
+Let's quickly [get started](./index.md) and learn how to manage embeddings in LanceDB. 
+
+## Bonus: As a developer, what you can create using embeddings?
+
+As a developer, you can create a variety of innovative applications using vector embeddings. Check out the following - 
+
+<div class="grid cards" markdown>
+
+-   __Chatbots__
+
+    ---
+
+    Develop chatbots that utilize embeddings to retrieve relevant context and generate coherent, contextually aware responses to user queries.
+
+    [:octicons-arrow-right-24: Check out examples](../examples/python_examples/chatbot.md)
+
+-   __Recommendation Systems__
+
+    ---
+
+    Develop systems that recommend content (such as articles, movies, or products) based on the similarity of keywords and descriptions, enhancing user experience.
+
+    [:octicons-arrow-right-24: Check out examples](../examples/python_examples/recommendersystem.md)
+
+-   __Vector Search__
+
+    ---
+
+    Build powerful applications that harness the full potential of semantic search, enabling them to retrieve relevant data quickly and effectively. 
+
+    [:octicons-arrow-right-24: Check out examples](../examples/python_examples/vector_search.md)
+
+-   __RAG Applications__
+
+    ---
+
+    Combine the strengths of large language models (LLMs) with retrieval-based approaches to create more useful applications.
+
+    [:octicons-arrow-right-24: Check out examples](../examples/python_examples/rag.md)
+
+-   __Many more examples__
+
+    ---
+
+    Explore applied examples available as Colab notebooks or Python scripts to integrate into your applications.
+
+    [:octicons-arrow-right-24: More](../examples/examples_python.md)
+
+</div>
+
+
+
+
+
+
+
+

docs/src/embeddings/variables_and_secrets.md

@@ -0,0 +1,53 @@
+# Variable and Secrets
+
+Most embedding configuration options are saved in the table's metadata. However,
+this isn't always appropriate. For example, API keys should never be stored in the
+metadata. Additionally, other configuration options might be best set at runtime,
+such as the `device` configuration that controls whether to use GPU or CPU for
+inference. If you hardcoded this to GPU, you wouldn't be able to run the code on
+a server without one.
+
+To handle these cases, you can set variables on the embedding registry and
+reference them in the embedding configuration. These variables will be available
+during the runtime of your program, but not saved in the table's metadata. When
+the table is loaded from a different process, the variables must be set again.
+
+To set a variable, use the `set_var()` / `setVar()` method on the embedding registry.
+To reference a variable, use the syntax `$env:VARIABLE_NAME`. If there is a default
+value, you can use the syntax `$env:VARIABLE_NAME:DEFAULT_VALUE`.
+
+## Using variables to set secrets
+
+Sensitive configuration, such as API keys, must either be set as environment
+variables or using variables on the embedding registry. If you pass in a hardcoded
+value, LanceDB will raise an error. Instead, if you want to set an API key via
+configuration, use a variable:
+
+=== "Python"
+
+    ```python
+    --8<-- "python/python/tests/docs/test_embeddings_optional.py:register_secret"
+    ```
+
+=== "Typescript"
+
+    ```typescript
+    --8<-- "nodejs/examples/embedding.test.ts:register_secret"
+    ```
+
+## Using variables to set the device parameter
+
+Many embedding functions that run locally have a `device` parameter that controls
+whether to use GPU or CPU for inference. Because not all computers have a GPU,
+it's helpful to be able to set the `device` parameter at runtime, rather than
+have it hard coded in the embedding configuration. To make it work even if the
+variable isn't set, you could provide a default value of `cpu` in the embedding
+configuration.
+
+Some embedding libraries even have a method to detect which devices are available,
+which could be used to dynamically set the device at runtime. For example, in Python
+you can check if a CUDA GPU is available using `torch.cuda.is_available()`.
+
+```python
+--8<-- "python/python/tests/docs/test_embeddings_optional.py:register_device"
+```

docs/src/examples/examples_python.md

@@ -1,17 +1,22 @@
-# Examples: Python
+# Overview : Python Examples 
 
-To help you get started, we provide some examples, projects and applications that use the LanceDB Python API. You can always find the latest examples in our [VectorDB Recipes](https://github.com/lancedb/vectordb-recipes) repository.
+To help you get started, we provide some examples, projects, and applications that use the LanceDB Python API. These examples are designed to get you right into the code with minimal introduction, enabling you to move from an idea to a proof of concept in minutes. 
 
-| Example | Interactive Envs | Scripts  |
-|-------- | ---------------- | ------   |
-| | | |
-| [Youtube transcript search bot](https://github.com/lancedb/vectordb-recipes/tree/main/examples/youtube_bot/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/youtube_bot/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/youtube_bot/main.py)|
-| [Langchain: Code Docs QA bot](https://github.com/lancedb/vectordb-recipes/tree/main/examples/Code-Documentation-QA-Bot/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/Code-Documentation-QA-Bot/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/Code-Documentation-QA-Bot/main.py) |
-| [AI Agents: Reducing Hallucination](https://github.com/lancedb/vectordb-recipes/tree/main/examples/reducing_hallucinations_ai_agents/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/reducing_hallucinations_ai_agents/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/reducing_hallucinations_ai_agents/main.py)|
-| [Multimodal CLIP: DiffusionDB](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_clip/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/multimodal_clip/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_clip/main.py)  |
-| [Multimodal CLIP: Youtube videos](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_video_search/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/multimodal_video_search/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>| [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_video_search/main.py)  |
-| [Movie Recommender](https://github.com/lancedb/vectordb-recipes/tree/main/examples/movie-recommender/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/movie-recommender/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> | [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/movie-recommender/main.py)  |
-| [Audio Search](https://github.com/lancedb/vectordb-recipes/tree/main/examples/audio_search/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/audio_search/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> | [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/audio_search/main.py)  |
-| [Multimodal Image + Text Search](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_search/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/multimodal_search/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> | [![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/examples/multimodal_search/main.py)  |
-| [Evaluating Prompts with Prompttools](https://github.com/lancedb/vectordb-recipes/tree/main/examples/prompttools-eval-prompts/) | <a href="https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/prompttools-eval-prompts/main.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a> |  |
+You can find the latest examples in our [VectorDB Recipes](https://github.com/lancedb/vectordb-recipes) repository. 
+
+**Introduction**
+
+Explore applied examples available as Colab notebooks or Python scripts to integrate into your applications. You can also checkout our blog posts related to the particular example for deeper understanding.
+
+| Explore                                         | Description                                                                                                                                                      |
+|----------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| [**Build from Scratch with LanceDB** 🛠️🚀](python_examples/build_from_scratch.md)               | Start building your **GenAI applications** from the **ground up** using **LanceDB's** efficient vector-based document retrieval capabilities! Get started quickly with a solid foundation.      |
+| [**Multimodal Search with LanceDB** 🤹‍♂️🔍](python_examples/multimodal.md)               | Combine **text** and **image queries** to find the most relevant results using **LanceDB’s multimodal** capabilities. Leverage the efficient vector-based similarity search.   |
+| [**RAG (Retrieval-Augmented Generation) with LanceDB** 🔓🧐](python_examples/rag.md) | Build RAG (Retrieval-Augmented Generation) with **LanceDB** for efficient **vector-based information retrieval** and more accurate responses from AI.                    |
+| [**Vector Search: Efficient Retrieval** 🔓👀](python_examples/vector_search.md) | Use **LanceDB's** vector search capabilities to perform efficient and accurate **similarity searches**, enabling rapid discovery and retrieval of relevant documents in Large datasets.                                                 |
+| [**Chatbot applications with LanceDB** 🤖](python_examples/chatbot.md)                | Create **chatbots** that retrieves relevant context for **coherent and context-aware replies**, enhancing user experience through advanced conversational AI. |
+| [**Evaluation: Assessing Text Performance with Precision** 📊💡](python_examples/evaluations.md) | Develop **evaluation** applications that allows you to input reference and candidate texts to **measure** their performance across various metrics.                                                      |
+| [**AI Agents: Intelligent Collaboration** 🤖](python_examples/aiagent.md)              | Enable **AI agents** to communicate and collaborate efficiently through dense vector representations, achieving shared goals seamlessly.                                         |
+| [**Recommender Systems: Personalized Discovery** 🍿📺](python_examples/recommendersystem.md)      | Deliver **personalized experiences** by efficiently storing and querying item embeddings with **LanceDB's** powerful vector database capabilities.                        |
+| **Miscellaneous Examples🌟** | Find other **unique examples** and **creative solutions** using **LanceDB**, showcasing the flexibility and broad applicability of the platform. |
 

docs/src/examples/index.md

@@ -8,9 +8,5 @@ LanceDB provides language APIs, allowing you to embed a database in your languag
 * 👾 [JavaScript](examples_js.md) examples
 * 🦀 Rust examples (coming soon)
 
-## Applications powered by LanceDB
-
-| Project Name                                        | Description                                                                                                          | Screenshot                                |
-|-----------------------------------------------------|----------------------------------------------------------------------------------------------------------------------|-------------------------------------------|
-| [YOLOExplorer](https://github.com/lancedb/yoloexplorer) | Iterate on your YOLO / CV datasets using SQL, Vector semantic search, and more within seconds                  | ![YOLOExplorer](https://github.com/lancedb/vectordb-recipes/assets/15766192/ae513a29-8f15-4e0b-99a1-ccd8272b6131) |
-| [Website Chatbot (Deployable Vercel Template)](https://github.com/lancedb/lancedb-vercel-chatbot) | Create a chatbot from the sitemap of any website/docs of your choice. Built using vectorDB serverless native javascript package. | ![Chatbot](../assets/vercel-template.gif)    |
+!!! tip "Hosted LanceDB"
+    If you want S3 cost-efficiency and local performance via a simple serverless API, checkout **LanceDB Cloud**. For private deployments, high performance at extreme scale, or if you have strict security requirements, talk to us about **LanceDB Enterprise**. [Learn more](https://docs.lancedb.com/)

docs/src/examples/python_examples/aiagent.md

@@ -0,0 +1,27 @@
+# AI Agents: Intelligent Collaboration🤖
+
+Think of a platform where AI Agents can seamlessly exchange information, coordinate over tasks, and achieve shared targets with great efficiency💻📈.
+
+## Vector-Based Coordination: The Technical Advantage
+Leveraging LanceDB's vector-based capabilities, we can enable **AI agents 🤖** to communicate and collaborate through dense vector representations. AI agents can exchange information, coordinate on a task or work towards a common goal, just by giving queries📝.
+
+| **AI Agents** | **Description** | **Links** |
+|:--------------|:----------------|:----------|
+| **AI Agents: Reducing Hallucinationt📊** | 🤖💡 **Reduce AI hallucinations** using Critique-Based Contexting! Learn by Simplifying and Automating tedious workflows by going through fitness trainer agent example.💪 | [![Github](../../assets/github.svg)][hullucination_github] <br>[![Open In Collab](../../assets/colab.svg)][hullucination_colab] <br>[![Python](../../assets/python.svg)][hullucination_python] <br>[![Ghost](../../assets/ghost.svg)][hullucination_ghost] |                                                                                                                                                                                                                                                
+| **AI Trends Searcher: CrewAI🔍️** | 🔍️ Learn about **CrewAI Agents** ! Utilize the features of CrewAI - Role-based Agents, Task Management, and Inter-agent Delegation ! Make AI agents work together to do tricky stuff 😺| [![Github](../../assets/github.svg)][trend_github] <br>[![Open In Collab](../../assets/colab.svg)][trend_colab] <br>[![Ghost](../../assets/ghost.svg)][trend_ghost] |                                                                                                                                                                                                                                                
+| **SuperAgent Autogen🤖** | 💻 AI interactions with the Super Agent! Integrating **Autogen**, **LanceDB**, **LangChain**, **LiteLLM**, and **Ollama** to create AI agent that excels in understanding and processing complex queries.🤖 | [![Github](../../assets/github.svg)][superagent_github] <br>[![Open In Collab](../../assets/colab.svg)][superagent_colab] |
+
+
+[hullucination_github]: https://github.com/lancedb/vectordb-recipes/blob/main/examples/reducing_hallucinations_ai_agents
+[hullucination_colab]: https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/reducing_hallucinations_ai_agents/main.ipynb
+[hullucination_python]: https://github.com/lancedb/vectordb-recipes/blob/main/examples/reducing_hallucinations_ai_agents/main.py
+[hullucination_ghost]: https://blog.lancedb.com/how-to-reduce-hallucinations-from-llm-powered-agents-using-long-term-memory-72f262c3cc1f/
+
+[trend_github]: https://github.com/lancedb/vectordb-recipes/blob/main/examples/AI-Trends-with-CrewAI
+[trend_colab]: https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/AI-Trends-with-CrewAI/CrewAI_AI_Trends.ipynb 
+[trend_ghost]: https://blog.lancedb.com/track-ai-trends-crewai-agents-rag/
+
+[superagent_github]: https://github.com/lancedb/vectordb-recipes/blob/main/examples/SuperAgent_Autogen
+[superagent_colab]: https://colab.research.google.com/github/lancedb/vectordb-recipes/blob/main/examples/SuperAgent_Autogen/main.ipynb
+
+

docs/src/examples/python_examples/build_from_scratch.md

@@ -0,0 +1,13 @@
+# **Build from Scratch with LanceDB 🛠️🚀**
+
+Start building your GenAI applications from the ground up using **LanceDB's** efficient vector-based document retrieval capabilities! 📑
+
+**Get Started in Minutes ⏱️**
+
+These examples provide a solid foundation for building your own GenAI applications using LanceDB. Jump from idea to **proof of concept** quickly with applied examples. Get started and see what you can create! 💻
+
+| **Build From Scratch**                     | **Description**                                                                                              | **Links**                                                                                                                                                                                                                                                                                                                                                                                                                |
+|:-------------------------------------------|:-------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| **Build RAG from Scratch🚀💻**             | 📝 Create a **Retrieval-Augmented Generation** (RAG) model from scratch using LanceDB.                       | [![GitHub](https://img.shields.io/badge/github-%23121011.svg?style=for-the-badge&logo=github&logoColor=white)](https://github.com/lancedb/vectordb-recipes/tree/main/tutorials/RAG-from-Scratch)<br>[![Open In Collab](https://colab.research.google.com/assets/colab-badge.svg)]()                                                                                                                                  |
+| **Local RAG from Scratch with Llama3🔥💡** | 🐫 Build a local RAG model using **Llama3** and **LanceDB** for fast and efficient text generation.          | [![GitHub](https://img.shields.io/badge/github-%23121011.svg?style=for-the-badge&logo=github&logoColor=white)](https://github.com/lancedb/vectordb-recipes/tree/main/tutorials/Local-RAG-from-Scratch)<br>[![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/blob/main/tutorials/Local-RAG-from-Scratch/rag.py)    |
+| **Multi-Head RAG from Scratch📚💻**        | 🤯 Develop a **Multi-Head RAG model** from scratch, enabling generation of text based on multiple documents. | [![GitHub](https://img.shields.io/badge/github-%23121011.svg?style=for-the-badge&logo=github&logoColor=white)](https://github.com/lancedb/vectordb-recipes/tree/main/tutorials/Multi-Head-RAG-from-Scratch)<br>[![Python](https://img.shields.io/badge/python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://github.com/lancedb/vectordb-recipes/tree/main/tutorials/Multi-Head-RAG-from-Scratch) |