Bump version: 0.21.0-beta.2 → 0.21.0

Prior to this commit, issuing drop_all_tables on a listing database with
an external manifest store would delete physical tables but leave
references behind in the manifest store. The table drop would succeed,
but subsequent creation of a table with the same name would fail with a
conflict.

With this patch, the external manifest store is updated to account for
the dropped tables so that dropped table names can be reused.

.bumpversion.cfg

@@ -1,12 +0,0 @@
-[bumpversion]
-current_version = 0.4.4
-commit = True
-message = Bump version: {current_version} → {new_version}
-tag = True
-tag_name = v{new_version}
-
-[bumpversion:file:node/package.json]
-
-[bumpversion:file:rust/ffi/node/Cargo.toml]
-
-[bumpversion:file:rust/vectordb/Cargo.toml]

.bumpversion.toml

@@ -0,0 +1,125 @@
+[tool.bumpversion]
+current_version = "0.18.0-beta.0"
+parse = """(?x)
+    (?P<major>0|[1-9]\\d*)\\.
+    (?P<minor>0|[1-9]\\d*)\\.
+    (?P<patch>0|[1-9]\\d*)
+    (?:-(?P<pre_l>[a-zA-Z-]+)\\.(?P<pre_n>0|[1-9]\\d*))?
+"""
+serialize = [
+    "{major}.{minor}.{patch}-{pre_l}.{pre_n}",
+    "{major}.{minor}.{patch}",
+]
+search = "{current_version}"
+replace = "{new_version}"
+regex = false
+ignore_missing_version = false
+ignore_missing_files = false
+tag = true
+sign_tags = false
+tag_name = "v{new_version}"
+tag_message = "Bump version: {current_version} → {new_version}"
+allow_dirty = true
+commit = true
+message = "Bump version: {current_version} → {new_version}"
+commit_args = ""
+
+# 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
+
+    # 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 = "node/package.json"
+replace = "\"version\": \"{new_version}\","
+search = "\"version\": \"{current_version}\","
+
+[[tool.bumpversion.files]]
+filename = "nodejs/package.json"
+replace = "\"version\": \"{new_version}\","
+search = "\"version\": \"{current_version}\","
+
+# nodejs binary packages
+[[tool.bumpversion.files]]
+glob = "nodejs/npm/*/package.json"
+replace = "\"version\": \"{new_version}\","
+search = "\"version\": \"{current_version}\","
+
+# vectodb node binary packages
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-darwin-arm64\": \"{new_version}\""
+search = "\"@lancedb/vectordb-darwin-arm64\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-darwin-x64\": \"{new_version}\""
+search = "\"@lancedb/vectordb-darwin-x64\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-linux-arm64-gnu\": \"{new_version}\""
+search = "\"@lancedb/vectordb-linux-arm64-gnu\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-linux-x64-gnu\": \"{new_version}\""
+search = "\"@lancedb/vectordb-linux-x64-gnu\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-linux-arm64-musl\": \"{new_version}\""
+search = "\"@lancedb/vectordb-linux-arm64-musl\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-linux-x64-musl\": \"{new_version}\""
+search = "\"@lancedb/vectordb-linux-x64-musl\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-win32-x64-msvc\": \"{new_version}\""
+search = "\"@lancedb/vectordb-win32-x64-msvc\": \"{current_version}\""
+
+[[tool.bumpversion.files]]
+glob = "node/package.json"
+replace = "\"@lancedb/vectordb-win32-arm64-msvc\": \"{new_version}\""
+search = "\"@lancedb/vectordb-win32-arm64-msvc\": \"{current_version}\""
+
+# Cargo files
+# ------------
+[[tool.bumpversion.files]]
+filename = "rust/ffi/node/Cargo.toml"
+replace = "\nversion = \"{new_version}\""
+search = "\nversion = \"{current_version}\""
+
+[[tool.bumpversion.files]]
+filename = "rust/lancedb/Cargo.toml"
+replace = "\nversion = \"{new_version}\""
+search = "\nversion = \"{current_version}\""
+
+[[tool.bumpversion.files]]
+filename = "nodejs/Cargo.toml"
+replace = "\nversion = \"{new_version}\""
+search = "\nversion = \"{current_version}\""

.cargo/config.toml

@@ -0,0 +1,47 @@
+[profile.release]
+lto = "fat"
+codegen-units = 1
+
+[profile.release-with-debug]
+inherits = "release"
+debug = true
+# Prioritize compile time over runtime performance
+codegen-units = 16
+lto = "thin"
+
+[target.'cfg(all())']
+rustflags = [
+    "-Wclippy::all",
+    "-Wclippy::style",
+    "-Wclippy::fallible_impl_from",
+    "-Wclippy::manual_let_else",
+    "-Wclippy::redundant_pub_crate",
+    "-Wclippy::string_add_assign",
+    "-Wclippy::string_add",
+    "-Wclippy::string_lit_as_bytes",
+    "-Wclippy::string_to_string",
+    "-Wclippy::use_self",
+    "-Dclippy::cargo",
+    "-Dclippy::dbg_macro",
+    # not too much we can do to avoid multiple crate versions
+    "-Aclippy::multiple-crate-versions",
+    "-Aclippy::wildcard_dependencies",
+]
+
+[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-apple-darwin]
+rustflags = ["-C", "target-cpu=apple-m1", "-C", "target-feature=+neon,+fp16,+fhm,+dotprod"]
+
+# Not all Windows systems have the C runtime installed, so this avoids library
+# 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/labeler.yml

@@ -0,0 +1,33 @@
+version: 1
+appendOnly: true
+# Labels are applied based on conventional commits standard
+# https://www.conventionalcommits.org/en/v1.0.0/
+# These labels are later used in release notes. See .github/release.yml
+labels:
+# If the PR title has an ! before the : it will be considered a breaking change
+# For example, `feat!: add new feature` will be considered a breaking change
+- label: breaking-change
+  title: "^[^:]+!:.*"
+- label: breaking-change
+  body: "BREAKING CHANGE"
+- label: enhancement
+  title: "^feat(\\(.+\\))?!?:.*"
+- label: bug
+  title: "^fix(\\(.+\\))?!?:.*"
+- label: documentation
+  title: "^docs(\\(.+\\))?!?:.*"
+- label: performance
+  title: "^perf(\\(.+\\))?!?:.*"
+- label: ci
+  title: "^ci(\\(.+\\))?!?:.*"
+- label: chore
+  title: "^(chore|test|build|style)(\\(.+\\))?!?:.*"
+- label: Python
+  files:
+    - "^python\\/.*"
+- label: Rust
+  files:
+    - "^rust\\/.*"
+- label: typescript
+  files:
+    - "^node\\/.*"

.github/release_notes.json

@@ -0,0 +1,41 @@
+{
+    "ignore_labels": ["chore"],
+    "pr_template": "- ${{TITLE}} by @${{AUTHOR}} in ${{URL}}",
+    "categories": [
+        {
+            "title": "## 🏆 Highlights",
+            "labels": ["highlight"]
+        },
+        {
+            "title": "## 🛠 Breaking Changes",
+            "labels": ["breaking-change"]
+        },
+        {
+            "title": "## ⚠️ Deprecations ",
+            "labels": ["deprecation"]
+        },
+        {
+            "title": "## 🎉 New Features",
+            "labels": ["enhancement"]
+        },
+        {
+            "title": "## 🐛 Bug Fixes",
+            "labels": ["bug"]
+        },
+        {
+            "title": "## 📚 Documentation",
+            "labels": ["documentation"]
+        },
+        {
+            "title": "## 🚀 Performance Improvements",
+            "labels": ["performance"]
+        },
+        {
+            "title": "## Other Changes"
+        },
+        {
+            "title": "## 🔧 Build and CI",
+            "labels": ["ci"]
+        }
+    ]
+}

.github/workflows/build_linux_wheel/action.yml

@@ -0,0 +1,58 @@
+# We create a composite action to be re-used both for testing and for releasing
+name: build-linux-wheel
+description: "Build a manylinux wheel for lance"
+inputs:
+  python-minor-version:
+    description: "8, 9, 10, 11, 12"
+    required: true
+  args:
+    description: "--release"
+    required: false
+    default: ""
+  arm-build:
+    description: "Build for arm64 instead of x86_64"
+    # Note: this does *not* mean the host is arm64, since we might be cross-compiling.
+    required: false
+    default: "false"
+  manylinux:
+    description: "The manylinux version to build for"
+    required: false
+    default: "2_17"
+runs:
+  using: "composite"
+  steps:
+    - name: CONFIRM ARM BUILD
+      shell: bash
+      run: |
+        echo "ARM BUILD: ${{ inputs.arm-build }}"
+    - name: Build x86_64 Manylinux wheel
+      if: ${{ inputs.arm-build == 'false' }}
+      uses: PyO3/maturin-action@v1
+      with:
+        command: build
+        working-directory: python
+        target: x86_64-unknown-linux-gnu
+        manylinux: ${{ inputs.manylinux }}
+        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 \
+            && unzip /tmp/protoc.zip -d /usr/local \
+            && rm /tmp/protoc.zip
+    - name: Build Arm Manylinux Wheel
+      if: ${{ inputs.arm-build == 'true' }}
+      uses: PyO3/maturin-action@v1
+      with:
+        command: build
+        working-directory: python
+        docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
+        target: aarch64-unknown-linux-gnu
+        manylinux: ${{ inputs.manylinux }}
+        args: ${{ inputs.args }}
+        before-script-linux: |
+          set -e
+          yum install -y openssl-devel 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

@@ -0,0 +1,26 @@
+# We create a composite action to be re-used both for testing and for releasing
+name: build_wheel
+description: "Build a lance wheel"
+inputs:
+  python-minor-version:
+    description: "8, 9, 10, 11"
+    required: true
+  args:
+    description: "--release"
+    required: false
+    default: ""
+runs:
+  using: "composite"
+  steps:
+    - name: Install macos dependency
+      shell: bash
+      run: |
+        brew install protobuf
+    - name: Build wheel
+      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

.github/workflows/build_windows_wheel/action.yml

@@ -0,0 +1,34 @@
+# We create a composite action to be re-used both for testing and for releasing
+name: build_wheel
+description: "Build a lance wheel"
+inputs:
+  python-minor-version:
+    description: "8, 9, 10, 11"
+    required: true
+  args:
+    description: "--release"
+    required: false
+    default: ""
+runs:
+  using: "composite"
+  steps:
+    - 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: Build wheel
+      uses: PyO3/maturin-action@v1
+      with:
+        command: build
+        args: ${{ inputs.args }}
+        docker-options: "-e PIP_EXTRA_INDEX_URL=https://pypi.fury.io/lancedb/"
+        working-directory: python
+    - uses: actions/upload-artifact@v4
+      with:
+        name: windows-wheels
+        path: python\target\wheels

.github/workflows/cargo-publish.yml

@@ -1,13 +1,20 @@
 name: Cargo Publish
 
 on:
-  release:
+  push:
-    types: [ published ]
+    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
 
 env:
   # This env var is used by Swatinem/rust-cache@v2 for the cache
   # key, so we set it to make sure it is always consistent.
   CARGO_TERM_COLOR: always
+  # Up-to-date compilers needed for fp16kernels.
+  CC: gcc-12
+  CXX: g++-12
 
 jobs:
   build:
@@ -16,7 +23,7 @@ jobs:
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: rust
@@ -26,4 +33,4 @@ jobs:
           sudo apt install -y protobuf-compiler libssl-dev
       - name: Publish the package
         run: |
-          cargo publish -p vectordb --all-features --token ${{ secrets.CARGO_REGISTRY_TOKEN }}
+          cargo publish -p lancedb --all-features --token ${{ secrets.CARGO_REGISTRY_TOKEN }}

.github/workflows/dev.yml

@@ -0,0 +1,81 @@
+name: PR Checks
+
+on:
+  pull_request_target:
+    types: [opened, edited, synchronize, reopened]
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  labeler:
+    permissions:
+      pull-requests: write
+    name: Label PR
+    runs-on: ubuntu-latest
+    steps:
+      - uses: srvaroa/labeler@master
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+  commitlint:
+    permissions:
+      pull-requests: write
+    name: Verify PR title / description conforms to semantic-release
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/setup-node@v3
+        with:
+          node-version: "18"
+      # These rules are disabled because Github will always ensure there
+      # is a blank line between the title and the body and Github will
+      # word wrap the description field to ensure a reasonable max line
+      # length.
+      - run: npm install @commitlint/config-conventional
+      - run: >
+          echo 'module.exports = {
+            "rules": {
+              "body-max-line-length": [0, "always", Infinity],
+              "footer-max-line-length": [0, "always", Infinity],
+              "body-leading-blank": [0, "always"]
+            }
+          }' > .commitlintrc.js
+      - run: npx commitlint --extends @commitlint/config-conventional --verbose <<< $COMMIT_MSG
+        env:
+          COMMIT_MSG: >
+            ${{ github.event.pull_request.title }}
+
+            ${{ github.event.pull_request.body }}
+      - if: failure()
+        uses: actions/github-script@v6
+        with:
+          script: |
+            const message = `**ACTION NEEDED**
+              
+              Lance follows the [Conventional Commits specification](https://www.conventionalcommits.org/en/v1.0.0/) for release automation.
+
+              The PR title and description are used as the merge commit message.\
+              Please update your PR title and description to match the specification.
+
+              For details on the error please inspect the "PR Title Check" action.
+              `
+            // Get list of current comments
+            const comments = await github.paginate(github.rest.issues.listComments, {
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number
+            });
+            // Check if this job already commented
+            for (const comment of comments) {
+              if (comment.body === message) {
+                return // Already commented
+              }
+            }
+            // Post the comment about Conventional Commits
+            github.rest.issues.createComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number,
+              body: message
+            })
+            core.setFailed(message)

.github/workflows/docs.yml

@@ -24,12 +24,16 @@ jobs:
     environment:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
-    runs-on: ubuntu-22.04
+    runs-on: buildjet-8vcpu-ubuntu-2204
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
+      - name: Install dependecies needed for ubuntu
+        run: |
+          sudo apt install -y protobuf-compiler libssl-dev
+          rustup update && rustup default
       - name: Set up Python
-        uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
           python-version: "3.10"
           cache: "pip"
@@ -37,12 +41,12 @@ jobs:
       - name: Build Python
         working-directory: python
         run: |
-          python -m pip install -e .
+          python -m pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .
-          python -m pip install -r ../docs/requirements.txt
+          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: ${{ matrix.node-version }}
+          node-version: 20
           cache: 'npm'
           cache-dependency-path: node/package-lock.json
       - uses: Swatinem/rust-cache@v2
@@ -61,16 +65,16 @@ jobs:
         working-directory: node
         run: |
           npx typedoc --plugin typedoc-plugin-markdown --out ../docs/src/javascript src/index.ts
-          cp ../docs/src/javascript.md ../docs/src/javascript/javascript.md
       - name: Build docs
+        working-directory: docs
         run: |
-          PYTHONPATH=. mkdocs build -f docs/mkdocs.yml
+          PYTHONPATH=. mkdocs build
       - 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

@@ -18,30 +18,38 @@ on:
 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.
-  RUSTFLAGS: "-C debuginfo=1"
+  RUSTFLAGS: "-C debuginfo=1 -C target-cpu=haswell -C target-feature=+f16c,+avx2,+fma"
   RUST_BACKTRACE: "1"
 
 jobs:
   test-python:
     name: Test doc python code
-    runs-on: ${{ matrix.os }}
+    runs-on: ubuntu-24.04
-    strategy:
-      matrix:
-        python-minor-version: [ "11" ]
-        os: ["ubuntu-22.04"]
     steps:
     - name: Checkout
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
+    - name: Print CPU capabilities
+      run: cat /proc/cpuinfo
+    - name: Install protobuf
+      run: |
+        sudo apt update
+        sudo apt install -y protobuf-compiler
+    - name: Install dependecies needed for ubuntu
+      run: |
+        sudo apt install -y libssl-dev
+        rustup update && rustup default
     - name: Set up Python
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
-        python-version: 3.${{ matrix.python-minor-version }}
+        python-version: 3.11
         cache: "pip"
         cache-dependency-path: "docs/test/requirements.txt"
+    - name: Rust cache
+      uses: swatinem/rust-cache@v2
     - 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
@@ -52,45 +60,49 @@ jobs:
         for d in *; do cd "$d"; echo "$d".py; python "$d".py; cd ..; done
   test-node:
     name: Test doc nodejs code
-    runs-on: ${{ matrix.os }}
+    runs-on: ubuntu-24.04
+    timeout-minutes: 60
     strategy:
-      matrix:
+      fail-fast: false
-        node-version: [ "18" ]
-        os: ["ubuntu-22.04"]
     steps:
     - name: Checkout
-      uses: actions/checkout@v3
+      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@v3
+      uses: actions/setup-node@v4
       with:
-        node-version: ${{ matrix.node-version }}
+        node-version: 20
+    - name: Install protobuf
+      run: |
+        sudo apt update
+        sudo apt install -y protobuf-compiler
     - name: Install dependecies needed for ubuntu
-      if: ${{ matrix.os == 'ubuntu-22.04' }}
       run: |
-        sudo apt install -y protobuf-compiler libssl-dev
+        sudo apt install -y libssl-dev
-    - name: Install node dependencies
+        rustup update && rustup default
-      run: |
-        cd docs/test
-        npm install 
     - name: Rust cache
       uses: swatinem/rust-cache@v2
-    - name: Install LanceDB
+    - name: Install node dependencies
       run: |
-        cd docs/test/node_modules/vectordb 
+        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
-        npm run tsc
+        cd ../docs
-    - name: Create test files
+        npm install
-      run: |
-        cd docs/test
-        node md_testing.js
     - name: Test
       env:
         LANCEDB_URI: ${{ secrets.LANCEDB_URI }}
         LANCEDB_DEV_API_KEY: ${{ secrets.LANCEDB_DEV_API_KEY }}
       run: |
-        cd docs/test/node
+        cd docs
-        for d in *; do cd "$d"; echo "$d".js; node "$d".js; cd ..; done
+        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: "1.79.0"
+          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: ${{ secrets.GPG_PASSPHRASE }}
+      - name: Install dependencies
+        run: |
+          sudo apt -y -qq update
+          sudo apt install -y protobuf-compiler libssl-dev pkg-config
+      - name: Download artifact
+        uses: actions/download-artifact@v4
+      - name: Copy native libs
+        run: |
+          mkdir -p ./core/target/classes/nativelib/darwin-aarch64 ./core/target/classes/nativelib/linux-aarch64
+          cp ../liblancedb_jni_darwin_aarch64.zip/liblancedb_jni.dylib ./core/target/classes/nativelib/darwin-aarch64/liblancedb_jni.dylib
+          cp ../liblancedb_jni_linux_aarch64.zip/liblancedb_jni.so ./core/target/classes/nativelib/linux-aarch64/liblancedb_jni.so
+      - name: Dry run
+        if: github.event_name == 'pull_request'
+        run: |
+          mvn --batch-mode -DskipTests package
+      - name: Set github
+        run: |
+          git config --global user.email "LanceDB Github Runner"
+          git config --global user.name "dev+gha@lancedb.com"
+      - name: Publish with Java 8
+        if: github.event_name == 'release'
+        run: |
+          echo "use-agent" >> ~/.gnupg/gpg.conf
+          echo "pinentry-mode loopback" >> ~/.gnupg/gpg.conf
+          export GPG_TTY=$(tty)
+          mvn --batch-mode -DskipTests -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

@@ -0,0 +1,113 @@
+name: Build and Run Java JNI Tests
+on:
+  push:
+    branches:
+      - main
+    paths:
+      - java/**
+  pull_request:
+    paths:
+      - java/**
+      - rust/**
+      - .github/workflows/java.yml
+env:
+  # This env var is used by Swatinem/rust-cache@v2 for the cache
+  # key, so we set it to make sure it is always consistent.
+  CARGO_TERM_COLOR: always
+  # 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"
+  RUST_BACKTRACE: "1"
+  # according to: https://matklad.github.io/2021/09/04/fast-rust-builds.html
+  # CI builds are faster with incremental disabled.
+  CARGO_INCREMENTAL: "0"
+  CARGO_BUILD_JOBS: "1"
+jobs:
+  linux-build-java-11:
+    runs-on: ubuntu-22.04
+    name: ubuntu-22.04 + Java 11
+    defaults:
+      run:
+        working-directory: ./java
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: java/core/lancedb-jni
+      - 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 11
+        uses: actions/setup-java@v4
+        with:
+          distribution: temurin
+          java-version: 11
+          cache: "maven"
+      - 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 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
+      - 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 \
+          -XX:+IgnoreUnrecognizedVMOptions \
+          --add-opens=java.base/java.lang=ALL-UNNAMED \
+          --add-opens=java.base/java.lang.invoke=ALL-UNNAMED \
+          --add-opens=java.base/java.lang.reflect=ALL-UNNAMED \
+          --add-opens=java.base/java.io=ALL-UNNAMED \
+          --add-opens=java.base/java.net=ALL-UNNAMED \
+          --add-opens=java.base/java.nio=ALL-UNNAMED \
+          --add-opens=java.base/java.util=ALL-UNNAMED \
+          --add-opens=java.base/java.util.concurrent=ALL-UNNAMED \
+          --add-opens=java.base/java.util.concurrent.atomic=ALL-UNNAMED \
+          --add-opens=java.base/jdk.internal.ref=ALL-UNNAMED \
+          --add-opens=java.base/sun.nio.ch=ALL-UNNAMED \
+          --add-opens=java.base/sun.nio.cs=ALL-UNNAMED \
+          --add-opens=java.base/sun.security.action=ALL-UNNAMED \
+          --add-opens=java.base/sun.util.calendar=ALL-UNNAMED \
+          --add-opens=java.security.jgss/sun.security.krb5=ALL-UNNAMED \
+          -Djdk.reflect.useDirectMethodHandle=false \
+          -Dio.netty.tryReflectionSetAccessible=true"
+          JAVA_HOME=$JAVA_17 mvn clean test
+  

.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

@@ -1,59 +1,102 @@
 name: Create release commit
 
+# This workflow increments versions, tags the version, and pushes it.
+# When a tag is pushed, another workflow is triggered that creates a GH release
+# and uploads the binaries. This workflow is only for creating the tag.
+
+# This script will enforce that a minor version is incremented if there are any
+# breaking changes since the last minor increment. However, it isn't able to
+# differentiate between breaking changes in Node versus Python. If you wish to
+# bypass this check, you can manually increment the version and push the tag.
 on:
   workflow_dispatch:
     inputs:
       dry_run:
         description: 'Dry run (create the local commit/tags but do not push it)'
         required: true
-        default: "false"
+        default: false
-        type: choice
+        type: boolean
-        options:
+      type:
-          - "true"
-          - "false"
-      part:
         description: 'What kind of release is this?'
         required: true
-        default: 'patch'
+        default: 'preview'
         type: choice
         options:
-          - patch
+          - preview
-          - minor
+          - stable
-          - major
+      python:
+        description: 'Make a Python release'
+        required: true
+        default: true
+        type: boolean
+      other:
+        description: 'Make a Node/Rust/Java release'
+        required: true
+        default: true
+        type: boolean
+      bump-minor:
+        description: 'Bump minor version'
+        required: true
+        default: false
+        type: boolean
 
 jobs:
-  bump-version:
+  make-release:
-    runs-on: ubuntu-latest
+    # Creates tag and GH release. The GH release will trigger the build and release jobs.
+    runs-on: ubuntu-24.04
+    permissions:
+      contents: write
     steps:
-      - name: Check out main
+      - name: Output Inputs
-        uses: actions/checkout@v3
+        run: echo "${{ toJSON(github.event.inputs) }}"
+      - uses: actions/checkout@v4
         with:
-          ref: main
-          persist-credentials: false
           fetch-depth: 0
           lfs: true
+          # It's important we use our token here, as the default token will NOT
+          # 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 3.10
+      - name: Bump Python version
-        uses: actions/setup-python@v4
+        if: ${{ inputs.python }}
-        with:
+        working-directory: python
-          python-version: "3.10"
+        env:
-      - name: Bump version, create tag and commit
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: |
-          pip install bump2version
+          # Need to get the commit before bumping the version, so we can
-          bumpversion --verbose ${{ inputs.part }}
+          # determine if there are breaking changes in the next step as well.
-      - name: Push new version and tag
+          echo "COMMIT_BEFORE_BUMP=$(git rev-parse HEAD)" >> $GITHUB_ENV
-        if: ${{ inputs.dry_run }} == "false"
+
+          pip install bump-my-version PyGithub packaging
+          bash ../ci/bump_version.sh ${{ inputs.type }} ${{ inputs.bump-minor }} python-v
+      - name: Bump Node/Rust version
+        if: ${{ inputs.other }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          pip install bump-my-version PyGithub packaging
+          bash ci/bump_version.sh ${{ inputs.type }} ${{ inputs.bump-minor }} v $COMMIT_BEFORE_BUMP
+      - name: Push new version tag
+        if: ${{ !inputs.dry_run }}
         uses: ad-m/github-push-action@master
         with:
+          # Need to use PAT here too to trigger next workflow. See comment above.
           github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
-          branch: main
+          branch: ${{ github.ref }}
           tags: true
       - uses: ./.github/workflows/update_package_lock
-        if: ${{ inputs.dry_run }} == "false"
+        if: ${{ !inputs.dry_run && inputs.other }}
         with:
-          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          github_token: ${{ secrets.GITHUB_TOKEN }}
-
+      - uses: ./.github/workflows/update_package_lock_nodejs
+        if: ${{ !inputs.dry_run && inputs.other }}
+        with:
+          github_token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/node.yml

@@ -20,31 +20,11 @@ env:
   # "1" means line tables only, which is useful for panic tracebacks.
   #
   # Use native CPU to accelerate tests if possible, especially for f16
-  RUSTFLAGS: "-C debuginfo=1 -C target-cpu=native -C target-feature=+f16c,+avx2,+fma"
+  # 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:
-  lint:
-    name: Lint
-    runs-on: ubuntu-22.04
-    defaults:
-      run:
-        shell: bash
-        working-directory: node
-    steps:
-    - uses: actions/checkout@v3
-      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: Lint
-      run: |
-        npm ci
-        npm run lint
   linux:
     name: Linux (Node ${{ matrix.node-version }})
     timeout-minutes: 30
@@ -57,7 +37,7 @@ jobs:
         shell: bash
         working-directory: node
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true
@@ -89,7 +69,7 @@ jobs:
         shell: bash
         working-directory: node
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true
@@ -127,8 +107,9 @@ jobs:
       AWS_ENDPOINT: http://localhost:4566
       # this one is for dynamodb
       DYNAMODB_ENDPOINT: http://localhost:4566
+      ALLOW_HTTP: true
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true

.github/workflows/nodejs.yml

@@ -28,8 +28,12 @@ jobs:
       run:
         shell: bash
         working-directory: nodejs
+    env:
+      # Need up-to-date compilers for kernels
+      CC: gcc-12
+      CXX: g++-12
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true
@@ -48,7 +52,10 @@ jobs:
         cargo fmt --all -- --check
         cargo clippy --all --all-features -- -D warnings
         npm ci
-        npm run lint
+        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
@@ -61,7 +68,7 @@ jobs:
         shell: bash
         working-directory: nodejs
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true
@@ -80,17 +87,46 @@ jobs:
       run: |
         npm ci
         npm run build
+    - name: Setup localstack
+      working-directory: .
+      run: docker compose up --detach --wait
     - name: Test
+      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; 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-13"
+    runs-on: "macos-14"
     defaults:
       run:
         shell: bash
         working-directory: nodejs
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
       with:
         fetch-depth: 0
         lfs: true
@@ -111,4 +147,3 @@ jobs:
     - name: Test
       run: |
         npm run test
-

.github/workflows/npm-publish.yml

@@ -1,11 +1,13 @@
 name: NPM Publish
 
 on:
-  release:
+  push:
-    types: [ published ]
+    tags:
+      - "v*"
 
 jobs:
   node:
+    name: vectordb Typescript
     runs-on: ubuntu-latest
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
@@ -15,11 +17,11 @@ jobs:
         working-directory: node
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - uses: actions/setup-node@v3
         with:
           node-version: 20
-          cache: 'npm'
+          cache: "npm"
           cache-dependency-path: node/package-lock.json
       - name: Install dependencies
         run: |
@@ -31,13 +33,14 @@ jobs:
           npm run tsc
           npm pack
       - name: Upload Linux Artifacts
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: node-package
           path: |
             node/vectordb-*.tgz
 
   node-macos:
+    name: vectordb ${{ matrix.config.arch }}
     strategy:
       matrix:
         config:
@@ -45,13 +48,13 @@ jobs:
             runner: macos-13
           - arch: aarch64-apple-darwin
             # xlarge is implicitly arm64.
-            runner: macos-13-xlarge
+            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@v3
+        uses: actions/checkout@v4
       - name: Install system dependencies
         run: brew install protobuf
       - name: Install npm dependencies
@@ -61,15 +64,45 @@ jobs:
       - name: Build MacOS native node modules
         run: bash ci/build_macos_artifacts.sh ${{ matrix.config.arch }}
       - name: Upload Darwin Artifacts
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
-          name: native-darwin
+          name: node-native-darwin-${{ matrix.config.arch }}
           path: |
             node/dist/lancedb-vectordb-darwin*.tgz
 
+  nodejs-macos:
+    name: lancedb ${{ matrix.config.arch }}
+    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:
+  node-linux-gnu:
-    name: node-linux (${{ matrix.config.arch}}-unknown-linux-gnu
+    name: vectordb (${{ 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')
@@ -80,21 +113,195 @@ jobs:
           - arch: x86_64
             runner: ubuntu-latest
           - arch: aarch64
-            runner: buildjet-4vcpu-ubuntu-2204-arm
+            # 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@v3
+        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 }}
+          bash ci/build_linux_artifacts.sh ${{ matrix.config.arch }} ${{ matrix.config.arch }}-unknown-linux-gnu
       - name: Upload Linux Artifacts
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
-          name: native-linux
+          name: node-native-linux-${{ matrix.config.arch }}-gnu
           path: |
             node/dist/lancedb-vectordb-linux*.tgz
 
+  node-linux-musl:
+    name: vectordb (${{ matrix.config.arch}}-unknown-linux-musl)
+    runs-on: ubuntu-latest
+    container: alpine:edge
+    # 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
+          - arch: aarch64
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install common dependencies
+        run: |
+          apk add protobuf-dev curl clang mold grep npm bash
+          curl --proto '=https' --tlsv1.3 -sSf https://raw.githubusercontent.com/rust-lang/rustup/refs/heads/master/rustup-init.sh | sh -s -- -y
+          echo "source $HOME/.cargo/env" >> saved_env
+          echo "export CC=clang" >> saved_env
+          echo "export RUSTFLAGS='-Ctarget-cpu=haswell -Ctarget-feature=-crt-static,+avx2,+fma,+f16c -Clinker=clang -Clink-arg=-fuse-ld=mold'" >> saved_env
+      - name: Configure aarch64 build
+        if: ${{ matrix.config.arch == 'aarch64' }}
+        run: |
+          source "$HOME/.cargo/env"
+          rustup target add aarch64-unknown-linux-musl
+          crt=$(realpath $(dirname $(rustup which rustc))/../lib/rustlib/aarch64-unknown-linux-musl/lib/self-contained)
+          sysroot_lib=/usr/aarch64-unknown-linux-musl/usr/lib
+          apk_url=https://dl-cdn.alpinelinux.org/alpine/latest-stable/main/aarch64/
+          curl -sSf $apk_url > apk_list
+          for pkg in gcc libgcc musl; do curl -sSf $apk_url$(cat apk_list | grep -oP '(?<=")'$pkg'-\d.*?(?=")') | tar zxf -; done
+          mkdir -p $sysroot_lib
+          echo 'GROUP ( libgcc_s.so.1 -lgcc )' > $sysroot_lib/libgcc_s.so
+          cp usr/lib/libgcc_s.so.1 $sysroot_lib
+          cp usr/lib/gcc/aarch64-alpine-linux-musl/*/libgcc.a $sysroot_lib
+          cp lib/ld-musl-aarch64.so.1 $sysroot_lib/libc.so
+          echo '!<arch>' > $sysroot_lib/libdl.a
+          (cd $crt && cp crti.o crtbeginS.o crtendS.o crtn.o -t $sysroot_lib)
+          echo "export CARGO_BUILD_TARGET=aarch64-unknown-linux-musl" >> saved_env
+          echo "export RUSTFLAGS='-Ctarget-cpu=apple-m1 -Ctarget-feature=-crt-static,+neon,+fp16,+fhm,+dotprod -Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=--target=aarch64-unknown-linux-musl -Clink-arg=--sysroot=/usr/aarch64-unknown-linux-musl -Clink-arg=-lc'" >> saved_env
+      - name: Build Linux Artifacts
+        run: |
+          source ./saved_env
+          bash ci/manylinux_node/build_vectordb.sh ${{ matrix.config.arch }} ${{ matrix.config.arch }}-unknown-linux-musl
+      - name: Upload Linux Artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: node-native-linux-${{ matrix.config.arch }}-musl
+          path: |
+            node/dist/lancedb-vectordb-linux*.tgz
+
+  nodejs-linux-gnu:
+    name: lancedb (${{ 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 }}-gnu
+          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: ${{ matrix.config.arch == 'x86_64' }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: nodejs-dist
+          path: |
+            nodejs/dist/*
+            !nodejs/dist/*.node
+
+  nodejs-linux-musl:
+    name: lancedb (${{ matrix.config.arch}}-unknown-linux-musl
+    runs-on: ubuntu-latest
+    container: alpine:edge
+    # 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
+          - arch: aarch64
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install common dependencies
+        run: |
+          apk add protobuf-dev curl clang mold grep npm bash openssl-dev openssl-libs-static
+          curl --proto '=https' --tlsv1.3 -sSf https://raw.githubusercontent.com/rust-lang/rustup/refs/heads/master/rustup-init.sh | sh -s -- -y
+          echo "source $HOME/.cargo/env" >> saved_env
+          echo "export CC=clang" >> saved_env
+          echo "export RUSTFLAGS='-Ctarget-cpu=haswell -Ctarget-feature=-crt-static,+avx2,+fma,+f16c -Clinker=clang -Clink-arg=-fuse-ld=mold'" >> saved_env
+          echo "export X86_64_UNKNOWN_LINUX_MUSL_OPENSSL_INCLUDE_DIR=/usr/include" >> saved_env
+          echo "export X86_64_UNKNOWN_LINUX_MUSL_OPENSSL_LIB_DIR=/usr/lib" >> saved_env
+      - name: Configure aarch64 build
+        if: ${{ matrix.config.arch == 'aarch64' }}
+        run: |
+          source "$HOME/.cargo/env"
+          rustup target add aarch64-unknown-linux-musl
+          crt=$(realpath $(dirname $(rustup which rustc))/../lib/rustlib/aarch64-unknown-linux-musl/lib/self-contained)
+          sysroot_lib=/usr/aarch64-unknown-linux-musl/usr/lib
+          apk_url=https://dl-cdn.alpinelinux.org/alpine/latest-stable/main/aarch64/
+          curl -sSf $apk_url > apk_list
+          for pkg in gcc libgcc musl openssl-dev openssl-libs-static; do curl -sSf $apk_url$(cat apk_list | grep -oP '(?<=")'$pkg'-\d.*?(?=")') | tar zxf -; done
+          mkdir -p $sysroot_lib
+          echo 'GROUP ( libgcc_s.so.1 -lgcc )' > $sysroot_lib/libgcc_s.so
+          cp usr/lib/libgcc_s.so.1 $sysroot_lib
+          cp usr/lib/gcc/aarch64-alpine-linux-musl/*/libgcc.a $sysroot_lib
+          cp lib/ld-musl-aarch64.so.1 $sysroot_lib/libc.so
+          echo '!<arch>' > $sysroot_lib/libdl.a
+          (cd $crt && cp crti.o crtbeginS.o crtendS.o crtn.o -t $sysroot_lib)
+          echo "export CARGO_BUILD_TARGET=aarch64-unknown-linux-musl" >> saved_env
+          echo "export RUSTFLAGS='-Ctarget-feature=-crt-static,+neon,+fp16,+fhm,+dotprod -Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=--target=aarch64-unknown-linux-musl -Clink-arg=--sysroot=/usr/aarch64-unknown-linux-musl -Clink-arg=-lc'" >> saved_env
+          echo "export AARCH64_UNKNOWN_LINUX_MUSL_OPENSSL_INCLUDE_DIR=$(realpath usr/include)" >> saved_env
+          echo "export AARCH64_UNKNOWN_LINUX_MUSL_OPENSSL_LIB_DIR=$(realpath usr/lib)" >> saved_env
+      - name: Build Linux Artifacts
+        run: |
+          source ./saved_env
+          bash ci/manylinux_node/build_lancedb.sh ${{ matrix.config.arch }}
+      - name: Upload Linux Artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: nodejs-native-linux-${{ matrix.config.arch }}-musl
+          path: |
+            nodejs/dist/*.node
+
   node-windows:
+    name: vectordb ${{ matrix.target }}
     runs-on: windows-2022
     # Only runs on tags that matches the make-release action
     if: startsWith(github.ref, 'refs/tags/v')
@@ -104,7 +311,7 @@ jobs:
         target: [x86_64-pc-windows-msvc]
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Install Protoc v21.12
         working-directory: C:\
         run: |
@@ -121,45 +328,337 @@ jobs:
       - name: Build Windows native node modules
         run: .\ci\build_windows_artifacts.ps1 ${{ matrix.target }}
       - name: Upload Windows Artifacts
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
-          name: native-windows
+          name: node-native-windows
           path: |
             node/dist/lancedb-vectordb-win32*.tgz
 
+  node-windows-arm64:
+    name: vectordb ${{ matrix.config.arch }}-pc-windows-msvc
+    # if: startsWith(github.ref, 'refs/tags/v')
+    runs-on: ubuntu-latest
+    container: alpine:edge
+    strategy:
+      fail-fast: false
+      matrix:
+        config:
+          # - arch: x86_64
+          - arch: aarch64
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install dependencies
+        run: |
+          apk add protobuf-dev curl clang lld llvm19 grep npm bash msitools sed
+          curl --proto '=https' --tlsv1.3 -sSf https://raw.githubusercontent.com/rust-lang/rustup/refs/heads/master/rustup-init.sh | sh -s -- -y
+          echo "source $HOME/.cargo/env" >> saved_env
+          echo "export CC=clang" >> saved_env
+          echo "export AR=llvm-ar" >> saved_env
+          source "$HOME/.cargo/env"
+          rustup target add ${{ matrix.config.arch }}-pc-windows-msvc
+          (mkdir -p sysroot && cd sysroot && sh ../ci/sysroot-${{ matrix.config.arch }}-pc-windows-msvc.sh)
+          echo "export C_INCLUDE_PATH=/usr/${{ matrix.config.arch }}-pc-windows-msvc/usr/include" >> saved_env
+          echo "export CARGO_BUILD_TARGET=${{ matrix.config.arch }}-pc-windows-msvc" >> saved_env
+      - name: Configure x86_64 build
+        if: ${{ matrix.config.arch == 'x86_64' }}
+        run: |
+          echo "export RUSTFLAGS='-Ctarget-cpu=haswell -Ctarget-feature=+crt-static,+avx2,+fma,+f16c -Clinker=lld -Clink-arg=/LIBPATH:/usr/x86_64-pc-windows-msvc/usr/lib'" >> saved_env
+      - name: Configure aarch64 build
+        if: ${{ matrix.config.arch == 'aarch64' }}
+        run: |
+          echo "export RUSTFLAGS='-Ctarget-feature=+crt-static,+neon,+fp16,+fhm,+dotprod -Clinker=lld -Clink-arg=/LIBPATH:/usr/aarch64-pc-windows-msvc/usr/lib -Clink-arg=arm64rt.lib'" >> saved_env
+      - name: Build Windows Artifacts
+        run: |
+          source ./saved_env
+          bash ci/manylinux_node/build_vectordb.sh ${{ matrix.config.arch }} ${{ matrix.config.arch }}-pc-windows-msvc
+      - name: Upload Windows Artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: node-native-windows-${{ matrix.config.arch }}
+          path: |
+            node/dist/lancedb-vectordb-win32*.tgz
+
+  nodejs-windows:
+    name: lancedb ${{ matrix.target }}
+    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
+
+  nodejs-windows-arm64:
+    name: lancedb ${{ matrix.config.arch }}-pc-windows-msvc
+    # Only runs on tags that matches the make-release action
+    # if: startsWith(github.ref, 'refs/tags/v')
+    runs-on: ubuntu-latest
+    container: alpine:edge
+    strategy:
+      fail-fast: false
+      matrix:
+        config:
+          # - arch: x86_64
+          - arch: aarch64
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install dependencies
+        run: |
+          apk add protobuf-dev curl clang lld llvm19 grep npm bash msitools sed
+          curl --proto '=https' --tlsv1.3 -sSf https://raw.githubusercontent.com/rust-lang/rustup/refs/heads/master/rustup-init.sh | sh -s -- -y
+          echo "source $HOME/.cargo/env" >> saved_env
+          echo "export CC=clang" >> saved_env
+          echo "export AR=llvm-ar" >> saved_env
+          source "$HOME/.cargo/env"
+          rustup target add ${{ matrix.config.arch }}-pc-windows-msvc
+          (mkdir -p sysroot && cd sysroot && sh ../ci/sysroot-${{ matrix.config.arch }}-pc-windows-msvc.sh)
+          echo "export C_INCLUDE_PATH=/usr/${{ matrix.config.arch }}-pc-windows-msvc/usr/include" >> saved_env
+          echo "export CARGO_BUILD_TARGET=${{ matrix.config.arch }}-pc-windows-msvc" >> saved_env
+          printf '#!/bin/sh\ncargo "$@"' > $HOME/.cargo/bin/cargo-xwin
+          chmod u+x $HOME/.cargo/bin/cargo-xwin
+      - name: Configure x86_64 build
+        if: ${{ matrix.config.arch == 'x86_64' }}
+        run: |
+          echo "export RUSTFLAGS='-Ctarget-cpu=haswell -Ctarget-feature=+crt-static,+avx2,+fma,+f16c -Clinker=lld -Clink-arg=/LIBPATH:/usr/x86_64-pc-windows-msvc/usr/lib'" >> saved_env
+      - name: Configure aarch64 build
+        if: ${{ matrix.config.arch == 'aarch64' }}
+        run: |
+          echo "export RUSTFLAGS='-Ctarget-feature=+crt-static,+neon,+fp16,+fhm,+dotprod -Clinker=lld -Clink-arg=/LIBPATH:/usr/aarch64-pc-windows-msvc/usr/lib -Clink-arg=arm64rt.lib'" >> saved_env
+      - name: Build Windows Artifacts
+        run: |
+          source ./saved_env
+          bash ci/manylinux_node/build_lancedb.sh ${{ matrix.config.arch }}
+      - name: Upload Windows Artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: nodejs-native-windows-${{ matrix.config.arch }}
+          path: |
+            nodejs/dist/*.node
+
   release:
-    needs: [node, node-macos, node-linux, node-windows]
+    name: vectordb NPM Publish
+    needs: [node, node-macos, node-linux-gnu, node-linux-musl, node-windows, node-windows-arm64]
     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@v3
+      - 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'
+          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 $filename
+            npm publish $PUBLISH_ARGS $filename
           done
+      - name: Notify Slack Action
+        uses: ravsamhq/notify-slack-action@2.3.0
+        if: ${{ always() }}
+        with:
+          status: ${{ job.status }}
+          notify_when: "failure"
+          notification_title: "{workflow} is failing"
+        env:
+          SLACK_WEBHOOK_URL: ${{ secrets.ACTION_MONITORING_SLACK }}
 
-  update-package-lock:
+  release-nodejs:
-    needs: [release]
+    name: lancedb NPM Publish
+    needs: [nodejs-macos, nodejs-linux-gnu, nodejs-linux-musl, nodejs-windows, nodejs-windows-arm64]
     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@v3
+        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
+      - name: Notify Slack Action
+        uses: ravsamhq/notify-slack-action@2.3.0
+        if: ${{ always() }}
+        with:
+          status: ${{ job.status }}
+          notify_when: "failure"
+          notification_title: "{workflow} is failing"
+        env:
+          SLACK_WEBHOOK_URL: ${{ secrets.ACTION_MONITORING_SLACK }}
+
+  update-package-lock:
+    if: startsWith(github.ref, 'refs/tags/v')
+    needs: [release]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
         with:
           ref: main
-          persist-credentials: false
+          token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
           fetch-depth: 0
           lfs: true
       - uses: ./.github/workflows/update_package_lock
         with:
-          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+
+  update-package-lock-nodejs:
+    if: startsWith(github.ref, 'refs/tags/v')
+    needs: [release-nodejs]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          ref: main
+          token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          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
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Extract version
+        id: extract_version
+        env:
+          GITHUB_REF: ${{ github.ref }}
+        run: |
+          set -e
+          echo "Extracting tag and version from $GITHUB_REF"
+          if [[ $GITHUB_REF =~ refs/tags/v(.*) ]]; then
+            VERSION=${BASH_REMATCH[1]}
+            TAG=v$VERSION
+            echo "tag=$TAG" >> $GITHUB_OUTPUT
+            echo "version=$VERSION" >> $GITHUB_OUTPUT
+          else
+            echo "Failed to extract version from $GITHUB_REF"
+            exit 1
+          fi
+          echo "Extracted version $VERSION from $GITHUB_REF"
+          if [[ $VERSION =~ beta ]]; then
+            echo "This is a beta release"
+
+            # Get last release (that is not this one)
+            FROM_TAG=$(git tag --sort='version:refname' \
+              | grep ^v \
+              | grep -vF "$TAG" \
+              | python ci/semver_sort.py v \
+              | tail -n 1)
+          else
+            echo "This is a stable release"
+            # Get last stable tag (ignore betas)
+            FROM_TAG=$(git tag --sort='version:refname' \
+              | grep ^v \
+              | grep -vF "$TAG" \
+              | grep -v beta \
+              | python ci/semver_sort.py v \
+              | tail -n 1)
+          fi
+          echo "Found from tag $FROM_TAG"
+          echo "from_tag=$FROM_TAG" >> $GITHUB_OUTPUT
+      - name: Create Release Notes
+        id: release_notes
+        uses: mikepenz/release-changelog-builder-action@v4
+        with:
+          configuration: .github/release_notes.json
+          toTag: ${{ steps.extract_version.outputs.tag }}
+          fromTag: ${{ steps.extract_version.outputs.from_tag }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      - name: Create GH release
+        uses: softprops/action-gh-release@v2
+        with:
+          prerelease: ${{ contains('beta', github.ref) }}
+          tag_name: ${{ steps.extract_version.outputs.tag }}
+          token: ${{ secrets.GITHUB_TOKEN }}
+          generate_release_notes: false
+          name: Node/Rust LanceDB v${{ steps.extract_version.outputs.version }}
+          body: ${{ steps.release_notes.outputs.changelog }}

.github/workflows/pypi-publish.yml

@@ -1,31 +1,166 @@
 name: PyPI Publish
 
 on:
-  release:
+  push:
-    types: [ published ]
+    tags:
+      - 'python-v*'
 
 jobs:
-  publish:
+  linux:
-    runs-on: ubuntu-latest
+    name: Python ${{ matrix.config.platform }} manylinux${{ matrix.config.manylinux }}
-    # Only runs on tags that matches the python-make-release action
+    timeout-minutes: 60
-    if: startsWith(github.ref, 'refs/tags/python-v')
+    strategy:
-    defaults:
+      matrix:
-      run:
+        config:
-        shell: bash
+          - platform: x86_64
-        working-directory: python
+            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_17"
+            extra_args: ""
+            # 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@v3
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.8"
+          python-version: 3.8
-      - name: Build distribution
+      - uses: ./.github/workflows/build_linux_wheel
-        run: |
-          ls -la
-          pip install wheel setuptools --upgrade
-          python setup.py sdist bdist_wheel
-      - name: Publish
-        uses: pypa/gh-action-pypi-publish@v1.8.5
         with:
-          password: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
+          python-minor-version: 8
-          packages-dir: python/dist
+          args: "--release --strip ${{ matrix.config.extra_args }}"
+          arm-build: ${{ matrix.config.platform == 'aarch64' }}
+          manylinux: ${{ matrix.config.manylinux }}
+      - uses: ./.github/workflows/upload_wheel
+        with:
+          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
+          fury_token: ${{ secrets.FURY_TOKEN }}
+  mac:
+    timeout-minutes: 60
+    runs-on: ${{ matrix.config.runner }}
+    strategy:
+      matrix:
+        config:
+          - target: x86_64-apple-darwin
+            runner: macos-13
+          - target: aarch64-apple-darwin
+            runner: macos-14
+    env:
+      MACOSX_DEPLOYMENT_TARGET: 10.15
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: 3.12
+      - uses: ./.github/workflows/build_mac_wheel
+        with:
+          python-minor-version: 8
+          args: "--release --strip --target ${{ matrix.config.target }} --features fp16kernels"
+      - uses: ./.github/workflows/upload_wheel
+        with:
+          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
+          fury_token: ${{ secrets.FURY_TOKEN }}
+  windows:
+    timeout-minutes: 60
+    runs-on: windows-latest
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          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
+        with:
+          pypi_token: ${{ secrets.LANCEDB_PYPI_API_TOKEN }}
+          fury_token: ${{ secrets.FURY_TOKEN }}
+  gh-release:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Extract version
+        id: extract_version
+        env:
+          GITHUB_REF: ${{ github.ref }}
+        run: |
+          set -e
+          echo "Extracting tag and version from $GITHUB_REF"
+          if [[ $GITHUB_REF =~ refs/tags/python-v(.*) ]]; then
+            VERSION=${BASH_REMATCH[1]}
+            TAG=python-v$VERSION
+            echo "tag=$TAG" >> $GITHUB_OUTPUT
+            echo "version=$VERSION" >> $GITHUB_OUTPUT
+          else
+            echo "Failed to extract version from $GITHUB_REF"
+            exit 1
+          fi
+          echo "Extracted version $VERSION from $GITHUB_REF"
+          if [[ $VERSION =~ beta ]]; then
+            echo "This is a beta release"
+
+            # Get last release (that is not this one)
+            FROM_TAG=$(git tag --sort='version:refname' \
+              | grep ^python-v \
+              | grep -vF "$TAG" \
+              | python ci/semver_sort.py python-v \
+              | tail -n 1)
+          else
+            echo "This is a stable release"
+            # Get last stable tag (ignore betas)
+            FROM_TAG=$(git tag --sort='version:refname' \
+              | grep ^python-v \
+              | grep -vF "$TAG" \
+              | grep -v beta \
+              | python ci/semver_sort.py python-v \
+              | tail -n 1)
+          fi
+          echo "Found from tag $FROM_TAG"
+          echo "from_tag=$FROM_TAG" >> $GITHUB_OUTPUT
+      - name: Create Python Release Notes
+        id: python_release_notes
+        uses: mikepenz/release-changelog-builder-action@v4
+        with:
+          configuration: .github/release_notes.json
+          toTag: ${{ steps.extract_version.outputs.tag }}
+          fromTag: ${{ steps.extract_version.outputs.from_tag }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      - name: Create Python GH release
+        uses: softprops/action-gh-release@v2
+        with:
+          prerelease: ${{ contains('beta', github.ref) }}
+          tag_name: ${{ steps.extract_version.outputs.tag }}
+          token: ${{ secrets.GITHUB_TOKEN }}
+          generate_release_notes: false
+          name: Python LanceDB v${{ steps.extract_version.outputs.version }}
+          body: ${{ steps.python_release_notes.outputs.changelog }}

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

@@ -1,56 +0,0 @@
-name: Python - Create release commit
-
-on:
-  workflow_dispatch:
-    inputs:
-      dry_run:
-        description: 'Dry run (create the local commit/tags but do not push it)'
-        required: true
-        default: "false"
-        type: choice
-        options:
-          - "true"
-          - "false"
-      part:
-        description: 'What kind of release is this?'
-        required: true
-        default: 'patch'
-        type: choice
-        options:
-          - patch
-          - minor
-          - major
-
-jobs:
-  bump-version:
-    runs-on: ubuntu-latest
-    steps:
-    - name: Check out main
-      uses: actions/checkout@v3
-      with:
-        ref: main
-        persist-credentials: false
-        fetch-depth: 0
-        lfs: true
-    - 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 3.10
-      uses: actions/setup-python@v4
-      with:
-        python-version: "3.10"
-    - name: Bump version, create tag and commit
-      working-directory: python
-      run: |
-        pip install bump2version
-        bumpversion --verbose ${{ inputs.part }}
-    - name: Push new version and tag
-      if: ${{ inputs.dry_run }} == "false"
-      uses: ad-m/github-push-action@master
-      with:
-        github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
-        branch: main
-        tags: true
-

.github/workflows/python.yml

@@ -14,49 +14,165 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  linux:
+  lint:
+    name: "Lint"
     timeout-minutes: 30
-    strategy:
-      matrix:
-        python-minor-version: [ "8", "9", "10", "11" ]
     runs-on: "ubuntu-22.04"
     defaults:
       run:
         shell: bash
         working-directory: python
     steps:
-    - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
       - name: Set up Python
-      uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
-        python-version: 3.${{ matrix.python-minor-version }}
+          python-version: "3.12"
-    - name: Install lancedb
+      - name: Install ruff
         run: |
-        pip install -e .[tests]
+          pip install ruff==0.9.9
-        pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
-        pip install pytest pytest-mock ruff
       - name: Format check
         run: ruff format --check .
       - name: Lint
-      run: ruff .
+        run: ruff check .
-    - name: Run tests
+
-      run: pytest -m "not slow" -x -v --durations=30 tests
+  type-check:
-    - name: doctest
+    name: "Type Check"
-      run: pytest --doctest-modules lancedb
+    timeout-minutes: 30
+    runs-on: "ubuntu-22.04"
+    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"
+      - 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: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: python
+      - name: Install
+        run: |
+          pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests,dev,embeddings]
+          pip install tantivy
+          pip install mlx
+      - name: Doctest
+        run: pytest --doctest-modules python/lancedb
+  linux:
+    name: "Linux: python-3.${{ matrix.python-minor-version }}"
+    timeout-minutes: 30
+    strategy:
+      matrix:
+        python-minor-version: ["9", "12"]
+    runs-on: "ubuntu-24.04"
+    defaults:
+      run:
+        shell: bash
+        working-directory: python
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          lfs: true
+      - name: Install protobuf
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: 3.${{ matrix.python-minor-version }}
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: python
+      - uses: ./.github/workflows/build_linux_wheel
+      - uses: ./.github/workflows/run_tests
+        with:
+          integration: true
+      # Make sure wheels are not included in the Rust cache
+      - name: Delete wheels
+        run: rm -rf target/wheels
   platform:
-    name: "Platform: ${{ matrix.config.name }}"
+    name: "Mac: ${{ matrix.config.name }}"
     timeout-minutes: 30
     strategy:
       matrix:
         config:
-          - name: x86 Mac
+          - name: x86
             runner: macos-13
-          - name: Arm Mac
+          - name: Arm
-            runner: macos-13-xlarge
+            runner: macos-14
-          - name: x86 Windows
+    runs-on: "${{ matrix.config.runner }}"
+    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"
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: python
+      - uses: ./.github/workflows/build_mac_wheel
+      - uses: ./.github/workflows/run_tests
+      # Make sure wheels are not included in the Rust cache
+      - name: Delete wheels
+        run: rm -rf target/wheels
+  windows:
+    name: "Windows: ${{ matrix.config.name }}"
+    timeout-minutes: 60
+    strategy:
+      matrix:
+        config:
+          - name: x86
             runner: windows-latest
     runs-on: "${{ matrix.config.runner }}"
     defaults:
@@ -64,44 +180,46 @@ jobs:
         shell: bash
         working-directory: python
     steps:
-    - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
       - name: Set up Python
-      uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
-        python-version: "3.11"
+          python-version: "3.12"
-    - name: Install lancedb
+      - uses: Swatinem/rust-cache@v2
-      run: |
+        with:
-        pip install -e .[tests]
+          workspaces: python
-        pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
+      - uses: ./.github/workflows/build_windows_wheel
-        pip install pytest pytest-mock
+      - uses: ./.github/workflows/run_tests
-    - name: Run tests
+      # Make sure wheels are not included in the Rust cache
-      run: pytest -m "not slow" -x -v --durations=30 tests
+      - name: Delete wheels
+        run: rm -rf target/wheels
   pydantic1x:
     timeout-minutes: 30
-    runs-on: "ubuntu-22.04"
+    runs-on: "ubuntu-24.04"
     defaults:
       run:
         shell: bash
         working-directory: python
     steps:
-    - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
+      - name: Install dependencies
+        run: |
+          sudo apt update
+          sudo apt install -y protobuf-compiler
       - name: Set up Python
-      uses: actions/setup-python@v4
+        uses: actions/setup-python@v5
         with:
           python-version: 3.9
       - name: Install lancedb
         run: |
           pip install "pydantic<2"
-        pip install -e .[tests]
+          pip install --extra-index-url https://pypi.fury.io/lancedb/ -e .[tests]
-        pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
+          pip install tantivy
-        pip install pytest pytest-mock
       - name: Run tests
-      run: pytest -m "not slow" -x -v --durations=30 tests
+        run: pytest -m "not slow and not s3_test" -x -v --durations=30 python/tests
-    - name: doctest
-      run: pytest --doctest-modules lancedb

.github/workflows/run_tests/action.yml

@@ -0,0 +1,31 @@
+name: run-tests
+
+description: "Install lance wheel and run unit tests"
+inputs:
+  python-minor-version:
+    required: true
+    description: "8 9 10 11 12"
+  integration:
+    required: false
+    description: "Run integration tests"
+    default: "false"
+runs:
+  using: "composite"
+  steps:
+    - name: Install lancedb
+      shell: bash
+      run: |
+        pip3 install --extra-index-url https://pypi.fury.io/lancedb/ $(ls target/wheels/lancedb-*.whl)[tests,dev]
+    - name: Setup localstack for integration tests
+      if: ${{ inputs.integration == 'true' }}
+      shell: bash
+      working-directory: .
+      run: docker compose up --detach --wait
+    - name: pytest (with integration)
+      shell: bash
+      if: ${{ inputs.integration == 'true' }}
+      run: pytest -m "not slow" -x -v --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

.github/workflows/rust.yml

@@ -22,17 +22,21 @@ 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: clang-18
+      CXX: clang++-18
     steps:
-    - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
@@ -46,16 +50,51 @@ jobs:
       - name: Run format
         run: cargo fmt --all -- --check
       - name: Run clippy
-      run: cargo clippy --all --all-features -- -D warnings
+        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: clang-18
+      CXX: clang++-18
     steps:
-    - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
@@ -64,24 +103,39 @@ jobs:
           workspaces: rust
       - name: Install dependencies
         run: |
-          sudo apt update
+          # 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
+      - 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
+        run: cargo build --all-features --tests --locked --examples
       - name: Run tests
-      run: cargo test --all-features
+        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-13-xlarge" ]
+        mac-runner: ["macos-13", "macos-14"]
     runs-on: "${{ matrix.mac-runner }}"
     defaults:
       run:
         shell: bash
         working-directory: rust
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           lfs: true
@@ -92,14 +146,19 @@ jobs:
           workspaces: rust
       - name: Install dependencies
         run: brew install protobuf
-      - name: Build
-        run: cargo build --all-features
       - name: Run tests
-        run: cargo test --all-features
+        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
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: Swatinem/rust-cache@v2
         with:
           workspaces: rust
@@ -115,5 +174,170 @@ jobs:
       - name: Run tests
         run: |
           $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
-          cargo build
+          cargo test --features remote --locked
-          cargo test
+
+  windows-arm64-cross:
+    # We cross compile in Node releases, so we want to make sure
+    # this can run successfully.
+    runs-on: ubuntu-latest
+    container: alpine:edge
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install dependencies (part 1)
+        run: |
+          set -e
+          apk add protobuf-dev curl clang lld llvm19 grep npm bash msitools sed
+      - name: Install rust
+        uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          target: aarch64-pc-windows-msvc
+      - name: Install dependencies (part 2)
+        run: |
+          set -e
+          mkdir -p sysroot
+          cd sysroot
+          sh ../ci/sysroot-aarch64-pc-windows-msvc.sh
+      - name: Check
+        env:
+          CC: clang
+          AR: llvm-ar
+          C_INCLUDE_PATH: /usr/aarch64-pc-windows-msvc/usr/include
+          CARGO_BUILD_TARGET: aarch64-pc-windows-msvc
+          RUSTFLAGS: -Ctarget-feature=+crt-static,+neon,+fp16,+fhm,+dotprod -Clinker=lld -Clink-arg=/LIBPATH:/usr/aarch64-pc-windows-msvc/usr/lib -Clink-arg=arm64rt.lib
+        run: |
+          source $HOME/.cargo/env
+          cargo check --features remote --locked
+
+  windows-arm64:
+    runs-on: windows-4x-arm
+    steps:
+      - name: Install Git
+        run: |
+          Invoke-WebRequest -Uri "https://github.com/git-for-windows/git/releases/download/v2.44.0.windows.1/Git-2.44.0-64-bit.exe" -OutFile "git-installer.exe"
+          Start-Process -FilePath "git-installer.exe" -ArgumentList "/VERYSILENT", "/NORESTART" -Wait
+        shell: powershell
+      - name: Add Git to PATH
+        run: |
+          Add-Content $env:GITHUB_PATH "C:\Program Files\Git\bin"
+          $env:Path = [System.Environment]::GetEnvironmentVariable("Path","Machine") + ";" + [System.Environment]::GetEnvironmentVariable("Path","User")
+        shell: powershell
+      - name: Configure Git symlinks
+        run: git config --global core.symlinks true
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.13"
+      - name: Install Visual Studio Build Tools
+        run: |
+          Invoke-WebRequest -Uri "https://aka.ms/vs/17/release/vs_buildtools.exe" -OutFile "vs_buildtools.exe"
+          Start-Process -FilePath "vs_buildtools.exe" -ArgumentList "--quiet", "--wait", "--norestart", "--nocache", `
+            "--installPath", "C:\BuildTools", `
+            "--add", "Microsoft.VisualStudio.Component.VC.Tools.ARM64", `
+            "--add", "Microsoft.VisualStudio.Component.VC.Tools.x86.x64", `
+            "--add", "Microsoft.VisualStudio.Component.Windows11SDK.22621", `
+            "--add", "Microsoft.VisualStudio.Component.VC.ATL", `
+            "--add", "Microsoft.VisualStudio.Component.VC.ATLMFC", `
+            "--add", "Microsoft.VisualStudio.Component.VC.Llvm.Clang" -Wait
+        shell: powershell
+      - name: Add Visual Studio Build Tools to PATH
+        run: |
+          $vsPath = "C:\BuildTools\VC\Tools\MSVC"
+          $latestVersion = (Get-ChildItem $vsPath | Sort-Object {[version]$_.Name} -Descending)[0].Name
+          Add-Content $env:GITHUB_PATH "C:\BuildTools\VC\Tools\MSVC\$latestVersion\bin\Hostx64\arm64"
+          Add-Content $env:GITHUB_PATH "C:\BuildTools\VC\Tools\MSVC\$latestVersion\bin\Hostx64\x64"
+          Add-Content $env:GITHUB_PATH "C:\Program Files (x86)\Windows Kits\10\bin\10.0.22621.0\arm64"
+          Add-Content $env:GITHUB_PATH "C:\Program Files (x86)\Windows Kits\10\bin\10.0.22621.0\x64"
+          Add-Content $env:GITHUB_PATH "C:\BuildTools\VC\Tools\Llvm\x64\bin"
+
+          # Add MSVC runtime libraries to LIB
+          $env:LIB = "C:\BuildTools\VC\Tools\MSVC\$latestVersion\lib\arm64;" +
+                     "C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22621.0\um\arm64;" +
+                     "C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22621.0\ucrt\arm64"
+          Add-Content $env:GITHUB_ENV "LIB=$env:LIB"
+
+          # Add INCLUDE paths
+          $env:INCLUDE = "C:\BuildTools\VC\Tools\MSVC\$latestVersion\include;" +
+                        "C:\Program Files (x86)\Windows Kits\10\Include\10.0.22621.0\ucrt;" +
+                        "C:\Program Files (x86)\Windows Kits\10\Include\10.0.22621.0\um;" +
+                        "C:\Program Files (x86)\Windows Kits\10\Include\10.0.22621.0\shared"
+          Add-Content $env:GITHUB_ENV "INCLUDE=$env:INCLUDE"
+        shell: powershell
+      - name: Install Rust
+        run: |
+          Invoke-WebRequest https://win.rustup.rs/x86_64 -OutFile rustup-init.exe
+          .\rustup-init.exe -y --default-host aarch64-pc-windows-msvc --default-toolchain 1.83.0
+        shell: powershell
+      - name: Add Rust to PATH
+        run: |
+          Add-Content $env:GITHUB_PATH "$env:USERPROFILE\.cargo\bin"
+        shell: powershell
+      - uses: Swatinem/rust-cache@v2
+        with:
+          workspaces: rust
+      - name: Install 7-Zip ARM
+        run: |
+          New-Item -Path 'C:\7zip' -ItemType Directory
+          Invoke-WebRequest https://7-zip.org/a/7z2408-arm64.exe -OutFile C:\7zip\7z-installer.exe
+          Start-Process -FilePath C:\7zip\7z-installer.exe -ArgumentList '/S' -Wait
+        shell: powershell
+      - name: Add 7-Zip to PATH
+        run: Add-Content $env:GITHUB_PATH "C:\Program Files\7-Zip"
+        shell: powershell
+      - name: Install Protoc v21.12
+        working-directory: C:\
+        run: |
+          if (Test-Path 'C:\protoc') {
+            Write-Host "Protoc directory exists, skipping installation"
+            return
+          }
+          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
+          & 'C:\Program Files\7-Zip\7z.exe' x protoc.zip
+        shell: powershell
+      - name: Add Protoc to PATH
+        run: Add-Content $env:GITHUB_PATH "C:\protoc\bin"
+        shell: powershell
+      - name: Run tests
+        run: |
+          $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
+          cargo test --target aarch64-pc-windows-msvc --features remote --locked
+
+  msrv:
+    # Check the minimum supported Rust version
+    name: MSRV Check - Rust v${{ 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_nodejs/action.yml

@@ -0,0 +1,33 @@
+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/update_package_lock_run.yml

@@ -8,7 +8,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
           ref: main
           persist-credentials: false

.github/workflows/update_package_lock_run_nodejs.yml

@@ -0,0 +1,19 @@
+name: Update NodeJs package-lock.json
+
+on:
+  workflow_dispatch:
+
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    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.LANCEDB_RELEASE_TOKEN }}

.github/workflows/upload_wheel/action.yml

@@ -0,0 +1,45 @@
+name: upload-wheel
+
+description: "Upload wheels to Pypi"
+inputs:
+  pypi_token:
+    required: true
+    description: "release token for the repo"
+  fury_token:
+    required: true
+    description: "release token for the fury repo"
+
+runs:
+  using: "composite"
+  steps:
+  - name: Install dependencies
+    shell: bash
+    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
+        echo "repo=fury" >> $GITHUB_OUTPUT
+      else
+        echo "repo=pypi" >> $GITHUB_OUTPUT
+      fi
+  - name: Publish to PyPI
+    shell: bash
+    env:
+      FURY_TOKEN: ${{ inputs.fury_token }}
+      PYPI_TOKEN: ${{ inputs.pypi_token }}
+    run: |
+      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/
+      else
+        twine upload --repository ${{ steps.choose_repo.outputs.repo }} \
+          --username __token__ \
+          --password $PYPI_TOKEN \
+          target/wheels/lancedb-*.whl
+      fi

.gitignore

@@ -4,11 +4,11 @@
 **/__pycache__
 .DS_Store
 venv
+.venv
 
 .vscode
-
+.zed
 rust/target
-rust/Cargo.lock
 
 site
 
@@ -22,6 +22,11 @@ python/dist
 
 **/.hypothesis
 
+# Compiled Dynamic libraries
+*.so
+*.dylib
+*.dll
+
 ## Javascript
 *.node
 **/node_modules
@@ -29,9 +34,10 @@ python/dist
 node/dist
 node/examples/**/package-lock.json
 node/examples/**/dist
+nodejs/lancedb/native*
 dist
 
 ## Rust
 target
 
-Cargo.lock
+**/sccache.log

.pre-commit-config.yaml

@@ -5,17 +5,23 @@ repos:
       - id: check-yaml
       - id: end-of-file-fixer
       - id: trailing-whitespace
--   repo: https://github.com/psf/black
-    rev: 22.12.0
-    hooks:
-    -   id: black
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.0.277
+    rev: v0.9.9
     hooks:
       - id: ruff
--   repo: https://github.com/pycqa/isort
+  # - repo: https://github.com/RobertCraigie/pyright-python
-    rev: 5.12.0
+  #   rev: v1.1.395
+  #   hooks:
+  #     - id: pyright
+  #       args: ["--project", "python"]
+  #       additional_dependencies: [pyarrow-stubs]
+  - repo: local
     hooks:
-    - id: isort
+      - id: local-biome-check
-      name: isort (python)
+        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/.*

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

@@ -1,37 +1,72 @@
 [workspace]
-members = ["rust/ffi/node", "rust/vectordb", "nodejs"]
+members = [
+    "rust/ffi/node",
+    "rust/lancedb",
+    "nodejs",
+    "python",
+    "java/core/lancedb-jni",
+]
 # Python package needs to be built by maturin.
 exclude = ["python"]
 resolver = "2"
 
 [workspace.package]
 edition = "2021"
-authors = ["Lance Devs <dev@lancedb.com>"]
+authors = ["LanceDB Devs <dev@lancedb.com>"]
 license = "Apache-2.0"
 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" = "=0.9.9", "features" = ["dynamodb"] }
+lance = { "version" = "=0.24.1", "features" = ["dynamodb"] }
-lance-index = { "version" = "=0.9.9" }
+lance-io = { version = "=0.24.1" }
-lance-linalg = { "version" = "=0.9.9" }
+lance-index = { version = "=0.24.1" }
-lance-testing = { "version" = "=0.9.9" }
+lance-linalg = { version = "=0.24.1" }
+lance-table = { version = "=0.24.1" }
+lance-testing = { version = "=0.24.1" }
+lance-datafusion = { version = "=0.24.1" }
+lance-encoding = { version = "=0.24.1" }
 # Note that this one does not include pyarrow
-arrow = { version = "49.0.0", optional = false }
+arrow = { version = "54.1", optional = false }
-arrow-array = "49.0"
+arrow-array = "54.1"
-arrow-data = "49.0"
+arrow-data = "54.1"
-arrow-ipc = "49.0"
+arrow-ipc = "54.1"
-arrow-ord = "49.0"
+arrow-ord = "54.1"
-arrow-schema = "49.0"
+arrow-schema = "54.1"
-arrow-arith = "49.0"
+arrow-arith = "54.1"
-arrow-cast = "49.0"
+arrow-cast = "54.1"
 async-trait = "0"
-chrono = "0.4.23"
+datafusion = { version = "45.0", default-features = false }
-half = { "version" = "=2.3.1", default-features = false, features = [
+datafusion-catalog = "45.0"
+datafusion-common = { version = "45.0", default-features = false }
+datafusion-execution = "45.0"
+datafusion-expr = "45.0"
+datafusion-physical-plan = "45.0"
+env_logger = "0.11"
+half = { "version" = "=2.4.1", default-features = false, features = [
     "num-traits",
 ] }
 futures = "0"
 log = "0.4"
-object_store = "0.9.0"
+moka = { version = "0.12", features = ["future"] }
-snafu = "0.7.4"
+object_store = "0.11.0"
+pin-project = "1.0.7"
+snafu = "0.8"
 url = "2"
 num-traits = "0.2"
+rand = "0.8"
+regex = "1.10"
+lazy_static = "1"
+semver = "1.0.25"
+
+# Temporary pins to work around downstream issues
+# https://github.com/apache/arrow-rs/commit/2fddf85afcd20110ce783ed5b4cdeb82293da30b
+chrono = "=0.4.39"
+# https://github.com/RustCrypto/formats/issues/1684
+base64ct = "=1.6.0"
+
+# Workaround for: https://github.com/eira-fransham/crunchy/issues/13
+crunchy = "=0.2.2"

README.md

@@ -1,15 +1,16 @@
 <div align="center">
 <p align="center">
 
-<img width="275" alt="LanceDB Logo" src="https://user-images.githubusercontent.com/917119/226205734-6063d87a-1ecc-45fe-85be-1dea6383a3d8.png">
+<img width="275" alt="LanceDB Logo" src="https://github.com/lancedb/lancedb/assets/5846846/37d7c7ad-c2fd-4f56-9f16-fffb0d17c73a">
 
-**Developer-friendly, serverless vector database for AI applications**
+**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>
-[![Medium](https://img.shields.io/badge/Medium-12100E?style=for-the-badge&logo=medium&logoColor=white)](https://blog.lancedb.com/) 
+[![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)
+[![Gurubase](https://img.shields.io/badge/Gurubase-Ask%20LanceDB%20Guru-006BFF?style=for-the-badge)](https://gurubase.io/g/lancedb)
 
 </p>
 
@@ -20,7 +21,7 @@
 
 <hr />
 
-LanceDB is an open-source database for vector-search built with persistent storage, which greatly simplifies retrevial, filtering and management of embeddings.
+LanceDB is an open-source database for vector-search built with persistent storage, which greatly simplifies retrieval, filtering and management of embeddings.
 
 The key features of LanceDB include:
 
@@ -36,7 +37,7 @@ The key features of LanceDB include:
 
 * GPU support in building vector index(*).
 
-* Ecosystem integrations with [LangChain 🦜️🔗](https://python.langchain.com/en/latest/modules/indexes/vectorstores/examples/lanecdb.html), [LlamaIndex 🦙](https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html), Apache-Arrow, Pandas, Polars, DuckDB and more on the way.
+* 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.
 
 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.
 
@@ -44,19 +45,24 @@ LanceDB's core is written in Rust 🦀 and is built using <a href="https://githu
 
 **Javascript**
 ```shell
-npm install vectordb
+npm install @lancedb/lancedb
 ```
 
 ```javascript
-const lancedb = require('vectordb');
+import * as lancedb from "@lancedb/lancedb";
-const db = await lancedb.connect('data/sample-lancedb');
 
-const table = await db.createTable('vectors',
+const db = await lancedb.connect("data/sample-lancedb");
-      [{ id: 1, vector: [0.1, 0.2], item: "foo", price: 10 },
+const table = await db.createTable("vectors", [
-       { id: 2, vector: [1.1, 1.2], item: "bar", price: 50 }])
+	{ id: 1, vector: [0.1, 0.2], item: "foo", price: 10 },
+	{ id: 2, vector: [1.1, 1.2], item: "bar", price: 50 },
+], {mode: 'overwrite'});
 
-const query = table.search([0.1, 0.3]).limit(2);
+
-const results = await query.execute();
+const query = table.vectorSearch([0.1, 0.3]).limit(2);
+const results = await query.toArray();
+
+// You can also search for rows by specific criteria without involving a vector search.
+const rowsByCriteria = await table.query().where("price >= 10").toArray();
 ```
 
 **Python**
@@ -76,5 +82,5 @@ result = table.search([100, 100]).limit(2).to_pandas()
 ```
 
 ## Blogs, Tutorials & Videos
-* 📈 <a href="https://blog.eto.ai/benchmarking-random-access-in-lance-ed690757a826">2000x better performance with Lance over Parquet</a>
+* 📈 <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>
+* 🤖 <a href="https://github.com/lancedb/vectordb-recipes/tree/main/examples/Youtube-Search-QA-Bot">Build a question and answer bot with LanceDB</a>

ci/build_linux_artifacts.sh

@@ -1,6 +1,7 @@
 #!/bin/bash
 set -e
 ARCH=${1:-x86_64}
+TARGET_TRIPLE=${2:-x86_64-unknown-linux-gnu}
 
 # 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.
@@ -13,7 +14,9 @@ docker build \
     .
 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
+    bash ci/manylinux_node/build_vectordb.sh $ARCH $TARGET_TRIPLE

ci/build_linux_artifacts_nodejs.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+set -e
+ARCH=${1:-x86_64}
+
+# We pass down the current user so that when we later mount the local files
+# into the container, the files are accessible by the current user.
+pushd ci/manylinux_node
+docker build \
+    -t lancedb-node-manylinux-$ARCH \
+    --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 $(pwd):/io -w /io \
+    --memory-swap=-1 \
+    lancedb-node-manylinux-$ARCH \
+    bash ci/manylinux_node/build_lancedb.sh $ARCH

ci/build_macos_artifacts_nodejs.sh

@@ -0,0 +1,34 @@
+# 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

@@ -3,6 +3,7 @@
 # Targets supported:
 # - x86_64-pc-windows-msvc
 # - i686-pc-windows-msvc
+# - aarch64-pc-windows-msvc
 
 function Prebuild-Rust {
     param (
@@ -31,7 +32,7 @@ function Build-NodeBinaries {
 
 $targets = $args[0]
 if (-not $targets) {
-    $targets = "x86_64-pc-windows-msvc"
+    $targets = "x86_64-pc-windows-msvc", "aarch64-pc-windows-msvc"
 }
 
 Write-Host "Building artifacts for targets: $targets"

ci/build_windows_artifacts_nodejs.ps1

@@ -0,0 +1,42 @@
+# 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
+# - aarch64-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", "aarch64-pc-windows-msvc"
+}
+
+Write-Host "Building artifacts for targets: $targets"
+foreach ($target in $targets) {
+    Prebuild-Rust $target
+    Build-NodeBinaries $target
+}

ci/bump_version.sh

@@ -0,0 +1,51 @@
+set -e
+
+RELEASE_TYPE=${1:-"stable"}
+BUMP_MINOR=${2:-false}
+TAG_PREFIX=${3:-"v"} # Such as "python-v"
+HEAD_SHA=${4:-$(git rev-parse HEAD)}
+
+readonly SELF_DIR=$(cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )
+
+PREV_TAG=$(git tag --sort='version:refname' | grep ^$TAG_PREFIX | python $SELF_DIR/semver_sort.py $TAG_PREFIX | tail -n 1)
+echo "Found previous tag $PREV_TAG"
+
+# Initially, we don't want to tag if we are doing stable, because we will bump
+# again later. See comment at end for why.
+if [[ "$RELEASE_TYPE" == 'stable' ]]; then 
+  BUMP_ARGS="--no-tag"
+fi
+
+# If last is stable and not bumping minor
+if [[ $PREV_TAG != *beta* ]]; then
+    if [[ "$BUMP_MINOR" != "false" ]]; then
+      # X.Y.Z -> X.(Y+1).0-beta.0
+      bump-my-version bump -vv $BUMP_ARGS minor
+    else
+      # X.Y.Z -> X.Y.(Z+1)-beta.0
+      bump-my-version bump -vv $BUMP_ARGS patch
+    fi
+else
+  if [[ "$BUMP_MINOR" != "false" ]]; then
+    # X.Y.Z-beta.N -> X.(Y+1).0-beta.0
+    bump-my-version bump -vv $BUMP_ARGS minor
+  else
+    # X.Y.Z-beta.N -> X.Y.Z-beta.(N+1)
+    bump-my-version bump -vv $BUMP_ARGS pre_n
+  fi
+fi
+
+# The above bump will always bump to a pre-release version. If we are releasing
+# a stable version, bump the pre-release level ("pre_l") to make it stable.
+if [[ $RELEASE_TYPE == 'stable' ]]; then
+  # X.Y.Z-beta.N -> X.Y.Z
+  bump-my-version bump -vv pre_l
+fi
+
+# Validate that we have incremented version appropriately for breaking changes
+NEW_TAG=$(git describe --tags --exact-match HEAD)
+NEW_VERSION=$(echo $NEW_TAG | sed "s/^$TAG_PREFIX//")
+LAST_STABLE_RELEASE=$(git tag --sort='version:refname' | grep ^$TAG_PREFIX | grep -v beta | grep -vF "$NEW_TAG" | python $SELF_DIR/semver_sort.py $TAG_PREFIX | tail -n 1)
+LAST_STABLE_VERSION=$(echo $LAST_STABLE_RELEASE | sed "s/^$TAG_PREFIX//")
+
+python $SELF_DIR/check_breaking_changes.py $LAST_STABLE_RELEASE $HEAD_SHA $LAST_STABLE_VERSION $NEW_VERSION

ci/check_breaking_changes.py

@@ -0,0 +1,35 @@
+"""
+Check whether there are any breaking changes in the PRs between the base and head commits.
+If there are, assert that we have incremented the minor version.
+"""
+import argparse
+import os
+from packaging.version import parse
+
+from github import Github
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("base")
+    parser.add_argument("head")
+    parser.add_argument("last_stable_version")
+    parser.add_argument("current_version")
+    args = parser.parse_args()
+
+    repo = Github(os.environ["GITHUB_TOKEN"]).get_repo(os.environ["GITHUB_REPOSITORY"])
+    commits = repo.compare(args.base, args.head).commits
+    prs = (pr for commit in commits for pr in commit.get_pulls())
+
+    for pr in prs:
+        if any(label.name == "breaking-change" for label in pr.labels):
+            print(f"Breaking change in PR: {pr.html_url}")
+            break
+    else:
+        print("No breaking changes found.")
+        exit(0)
+    
+    last_stable_version = parse(args.last_stable_version)
+    current_version = parse(args.current_version)
+    if current_version.minor <= last_stable_version.minor:
+        print("Minor version is not greater than the last stable version.")
+        exit(1)

ci/manylinux_node/Dockerfile

@@ -4,7 +4,7 @@
 # range of linux distributions.
 ARG ARCH=x86_64
 
-FROM quay.io/pypa/manylinux2014_${ARCH}
+FROM quay.io/pypa/manylinux_2_28_${ARCH}
 
 ARG ARCH=x86_64
 ARG DOCKER_USER=default_user
@@ -18,8 +18,8 @@ COPY install_protobuf.sh install_protobuf.sh
 RUN ./install_protobuf.sh ${ARCH}
 
 ENV DOCKER_USER=${DOCKER_USER}
-# Create a group and user
+# Create a group and user, but only if it doesn't exist
-RUN echo ${ARCH} && adduser --user-group --create-home --uid ${DOCKER_USER} build_user
+RUN echo ${ARCH} && id -u ${DOCKER_USER} >/dev/null 2>&1 || 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.

ci/manylinux_node/build_lancedb.sh

@@ -0,0 +1,19 @@
+#!/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
+
+#Alpine doesn't have .bashrc
+FILE=$HOME/.bashrc && test -f $FILE && source $FILE
+
+cd nodejs
+npm ci
+npm run build-release

ci/manylinux_node/build_vectordb.sh

@@ -2,6 +2,7 @@
 # Builds the node module for manylinux. Invoked by ci/build_linux_artifacts.sh.
 set -e
 ARCH=${1:-x86_64}
+TARGET_TRIPLE=${2:-x86_64-unknown-linux-gnu}
 
 if [ "$ARCH" = "x86_64" ]; then
     export OPENSSL_LIB_DIR=/usr/local/lib64/
@@ -11,9 +12,10 @@ fi
 export OPENSSL_STATIC=1
 export OPENSSL_INCLUDE_DIR=/usr/local/include/openssl
 
-source $HOME/.bashrc
+#Alpine doesn't have .bashrc
+FILE=$HOME/.bashrc && test -f $FILE && source $FILE
 
 cd node
 npm ci
 npm run build-release
-npm run pack-build
+npm run pack-build -- -t $TARGET_TRIPLE

ci/manylinux_node/install_openssl.sh

@@ -6,7 +6,7 @@
 # /usr/bin/ld: failed to set dynamic section sizes: Bad value
 set -e
 
-git clone -b OpenSSL_1_1_1u \
+git clone -b OpenSSL_1_1_1v \
     --single-branch \
     https://github.com/openssl/openssl.git
 

ci/manylinux_node/prepare_manylinux_node.sh

@@ -8,7 +8,7 @@ install_node() {
 
     source "$HOME"/.bashrc
 
-    nvm install --no-progress 16
+    nvm install --no-progress 18
 }
 
 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/semver_sort.py

@@ -0,0 +1,35 @@
+"""
+Takes a list of semver strings and sorts them in ascending order.
+"""
+
+import sys
+from packaging.version import parse, InvalidVersion
+
+if __name__ == "__main__":
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument("prefix", default="v")
+    args = parser.parse_args()
+
+    # Read the input from stdin
+    lines = sys.stdin.readlines()
+
+    # Parse the versions
+    versions = []
+    for line in lines:
+        line = line.strip()
+        try:
+            version_str = line.removeprefix(args.prefix)
+            version = parse(version_str)
+        except InvalidVersion:
+            # There are old tags that don't follow the semver format
+            print(f"Invalid version: {line}", file=sys.stderr)
+            continue
+        versions.append((line, version))
+
+    # Sort the versions
+    versions.sort(key=lambda x: x[1])
+
+    # Print the sorted versions as original strings
+    for line, _ in versions:
+        print(line)

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/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")

docker-compose.yml

@@ -1,18 +1,18 @@
 version: "3.9"
 services:
   localstack:
-    image: localstack/localstack:0.14
+    image: localstack/localstack:3.3
     ports:
       - 4566:4566
     environment:
-      - SERVICES=s3,dynamodb
+      - SERVICES=s3,dynamodb,kms
       - DEBUG=1
       - LS_LOG=trace
       - DOCKER_HOST=unix:///var/run/docker.sock
       - AWS_ACCESS_KEY_ID=ACCESSKEY
       - AWS_SECRET_ACCESS_KEY=SECRETKEY
     healthcheck:
-      test: [ "CMD", "curl", "-f", "http://localhost:4566/health" ]
+      test: [ "CMD", "curl", "-s", "http://localhost:4566/_localstack/health" ]
       interval: 5s
       retries: 3
       start_period: 10s

dockerfiles/Dockerfile

@@ -0,0 +1,27 @@
+#Simple base dockerfile that supports basic dependencies required to run lance with FTS and Hybrid Search
+#Usage docker build -t lancedb:latest -f Dockerfile .
+FROM python:3.10-slim-buster
+
+# Install Rust
+RUN apt-get update && apt-get install -y curl build-essential && \
+  curl https://sh.rustup.rs -sSf | sh -s -- -y
+
+# Set the environment variable for Rust
+ENV PATH="/root/.cargo/bin:${PATH}"
+
+# Install protobuf compiler
+RUN apt-get install -y protobuf-compiler && \
+  apt-get clean && \
+  rm -rf /var/lib/apt/lists/*
+
+RUN apt-get -y update &&\
+  apt-get -y upgrade && \
+  apt-get -y install git
+
+
+# Verify installations
+RUN python --version && \
+  rustc --version && \
+  protoc --version
+
+RUN pip install tantivy lancedb

docs/README.md

@@ -9,27 +9,81 @@ unreleased features.
 ## Building the docs
 
 ### Setup
-1. Install LanceDB. From LanceDB repo root: `pip install -e python`
+1. Install LanceDB Python. See setup in [Python contributing guide](../python/CONTRIBUTING.md).
-2. Install dependencies. From LanceDB repo root: `pip install -r docs/requirements.txt`
+   Run `make develop` to install the Python package.
-3. Make sure you have node and npm setup
+2. Install documentation dependencies. From LanceDB repo root: `pip install -r docs/requirements.txt`
-4. Make sure protobuf and libssl are installed
 
-### Building node module and create markdown files
+### Preview the docs
 
-See [Javascript docs README](./src/javascript/README.md)
+```shell
-
-### 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
 cd docs
 mkdocs serve
 ```
+
+If you want to just generate the HTML files:
+
+```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"
@@ -26,16 +29,18 @@ theme:
     - content.code.copy
     - content.tabs.link
     - content.action.edit
+    - content.tooltips
     - toc.follow
-    # - toc.integrate
     - navigation.top
     - navigation.tabs
     - navigation.tabs.sticky
     - navigation.footer
     - navigation.tracking
     - navigation.instant
+    - content.footnote.tooltips
   icon:
     repo: fontawesome/brands/github
+    annotation: material/arrow-right-circle
   custom_dir: overrides
 
 plugins:
@@ -47,32 +52,55 @@ plugins:
           paths: [../python]
           options:
             docstring_style: numpy
-          heading_level: 4
+            heading_level: 3
             show_source: true
             show_symbol_type_in_heading: true
             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
 
 markdown_extensions:
   - admonition
   - footnotes
+  - pymdownx.critic
+  - pymdownx.caret
+  - pymdownx.keys
+  - pymdownx.mark
+  - pymdownx.tilde
   - pymdownx.details
   - pymdownx.highlight:
       anchor_linenums: true
       line_spans: __span
       pygments_lang_class: true
   - pymdownx.inlinehilite
-- pymdownx.snippets
+  - pymdownx.snippets:
+      base_path: ..
+      dedent_subsections: true
   - pymdownx.superfences
   - 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:
@@ -80,23 +108,81 @@ 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
+          - 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:
+              - Choosing right query type: guides/tuning_retrievers/1_query_types.md
+              - 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
-    - Explicit management: embeddings/embedding_explicit.md
+          - Get Started: embeddings/index.md
-    - Implicit management: embeddings/embedding_functions.md
+          - Embedding functions: embeddings/embedding_functions.md
-    - Available Functions: embeddings/default_embedding_functions.md
+          - Available models:
-    - Custom Embedding Functions: embeddings/api.md
+              - 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:
@@ -104,20 +190,31 @@ nav:
           - Pandas and PyArrow: python/pandas_and_pyarrow.md
           - Polars: python/polars_arrow.md
           - DuckDB: python/duckdb.md
-    - LangChain 🔗: https://python.langchain.com/en/latest/modules/indexes/vectorstores/examples/lancedb.html
+          - LangChain:
-    - LangChain JS/TS 🔗: https://js.langchain.com/docs/modules/data_connection/vectorstores/integrations/lancedb
+              - LangChain 🔗: integrations/langchain.md
-    - LlamaIndex 🦙: https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html
+              - 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
           - Pydantic: python/pydantic.md
           - Voxel51: integrations/voxel51.md
           - PromptTools: integrations/prompttools.md
+          - dlt: integrations/dlt.md
+          - phidata: integrations/phidata.md
       - 🎯 Examples:
           - Overview: examples/index.md
           - 🐍 Python:
               - Overview: examples/examples_python.md
-      - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
+              - Build From Scratch: examples/python_examples/build_from_scratch.md
-      - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
+              - Multimodal: examples/python_examples/multimodal.md
-      - Multimodal search using CLIP: notebooks/multimodal_search.ipynb
+              - Rag: examples/python_examples/rag.md
-      - Example - Calculate CLIP Embeddings with Roboflow Inference: examples/image_embeddings_roboflow.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:
@@ -125,35 +222,101 @@ nav:
               - 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
-  - 🔧 CLI & Config: cli_config.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: javascript/javascript.md
+          - 👾 JavaScript (vectordb): 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
+          - FAQs: cloud/cloud_faq.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
       - 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
+      - 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:
+          - Choosing right query type: guides/tuning_retrievers/1_query_types.md
+          - 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
-  - Explicit management: embeddings/embedding_explicit.md
+      - Get Started: embeddings/index.md
-  - Implicit management: embeddings/embedding_functions.md
+      - Embedding functions: embeddings/embedding_functions.md
-  - Available Functions: embeddings/default_embedding_functions.md
+      - Available models:
-  - Custom Embedding Functions: embeddings/api.md
+          - 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:
@@ -161,35 +324,69 @@ nav:
       - Pandas and PyArrow: python/pandas_and_pyarrow.md
       - Polars: python/polars_arrow.md
       - DuckDB: python/duckdb.md
-  - LangChain 🦜️🔗↗: https://python.langchain.com/en/latest/modules/indexes/vectorstores/examples/lancedb.html
+      - LangChain 🦜️🔗↗: integrations/langchain.md
-  - LangChain.js 🦜️🔗↗: https://js.langchain.com/docs/modules/data_connection/vectorstores/integrations/lancedb
+      - LangChain.js 🦜️🔗↗: https://js.langchain.com/docs/integrations/vectorstores/lancedb
-  - LlamaIndex 🦙↗: https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/LanceDBIndexDemo.html
+      - LlamaIndex 🦙↗: integrations/llamaIndex.md
       - Pydantic: python/pydantic.md
       - Voxel51: integrations/voxel51.md
       - PromptTools: integrations/prompttools.md
-- Python examples:
+      - dlt: integrations/dlt.md
+      - phidata: integrations/phidata.md
+  - Examples:
       - examples/index.md
-  - YouTube Transcript Search: notebooks/youtube_transcript_search.ipynb
+      - 🐍 Python:
-  - Documentation QA Bot using LangChain: notebooks/code_qa_bot.ipynb
+          - Overview: examples/examples_python.md
-  - Multimodal search using CLIP: notebooks/multimodal_search.ipynb
+          - 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 examples:
+      - 👾 JavaScript:
           - Overview: examples/examples_js.md
+          - Serverless Website Chatbot: examples/serverless_website_chatbot.md
           - YouTube Transcript Search: 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
+      - 🦀 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: javascript/javascript.md
+      - Javascript (vectordb): javascript/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
+      - FAQs: cloud/cloud_faq.md
 
 extra_css:
   - styles/global.css
   - styles/extra.css
 
+extra_javascript:
+  - "extra_js/init_ask_ai_widget.js"
+  - "extra_js/reo.js"
+
 extra:
   analytics:
     provider: google
     property: G-B7NFM40W74
+  social:
+    - icon: fontawesome/brands/github
+      link: https://github.com/lancedb/lancedb
+    - icon: fontawesome/brands/x-twitter
+      link: https://twitter.com/lancedb
+    - icon: fontawesome/brands/linkedin
+      link: https://www.linkedin.com/company/lancedb

docs/openapi.yml

@@ -0,0 +1,513 @@
+openapi: 3.1.0
+info:
+  version: 1.0.0
+  title: LanceDB Cloud API
+  description: |
+    LanceDB Cloud API is a RESTful API that allows users to access and modify data stored in LanceDB Cloud.
+    Table actions are considered temporary resource creations and all use POST method.
+  contact:
+    name: LanceDB support
+    url: https://lancedb.com
+    email: contact@lancedb.com
+
+servers:
+  - url: https://{db}.{region}.api.lancedb.com
+    description: LanceDB Cloud REST endpoint.
+    variables:
+      db:
+        default: ""
+        description: the name of DB
+      region:
+        default: "us-east-1"
+        description: the service region of the DB
+
+security:
+  - key_auth: []
+
+components:
+  securitySchemes:
+    key_auth:
+      name: x-api-key
+      type: apiKey
+      in: header
+  parameters:
+    table_name:
+      name: name
+      in: path
+      description: name of the table
+      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
+      content:
+        text/plain:
+          schema:
+            type: string
+    not_found:
+      description: Not found
+      content:
+        text/plain:
+          schema:
+            type: string
+    unauthorized:
+      description: Unauthorized
+      content:
+        text/plain:
+          schema:
+            type: string
+  requestBodies:
+    arrow_stream_buffer:
+      description: Arrow IPC stream buffer
+      required: true
+      content:
+        application/vnd.apache.arrow.stream:
+          schema:
+            type: string
+            format: binary
+
+paths:
+  /v1/table/:
+    get:
+      description: List tables, optionally, with pagination.
+      tags:
+        - Tables
+      summary: List Tables
+      operationId: listTables
+      parameters:
+        - name: limit
+          in: query
+          description: Limits the number of items to return.
+          schema:
+            type: integer
+        - name: page_token
+          in: query
+          description: Specifies the starting position of the next query
+          schema:
+            type: string
+      responses:
+        "200":
+          description: Successfully returned a list of tables in the DB
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  tables:
+                    type: array
+                    items:
+                      type: string
+                  page_token:
+                    type: string
+
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+
+  /v1/table/{name}/create/:
+    post:
+      description: Create a new table
+      summary: Create a new table
+      operationId: createTable
+      tags:
+        - Tables
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        $ref: "#/components/requestBodies/arrow_stream_buffer"
+      responses:
+        "200":
+          description: Table successfully created
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+
+  /v1/table/{name}/query/:
+    post:
+      description: Vector Query
+      url: https://{db-uri}.{aws-region}.api.lancedb.com/v1/table/{name}/query/
+      tags:
+        - Data
+      summary: Vector Query
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              properties:
+                vector:
+                  type: FixedSizeList
+                  description: |
+                    The targetted vector to search for. Required.
+                vector_column:
+                  type: string
+                  description: |
+                    The column to query, it can be inferred from the schema if there is only one vector column.
+                prefilter:
+                  type: boolean
+                  description: |
+                    Whether to prefilter the data. Optional.
+                k:
+                  type: integer
+                  description: |
+                    The number of search results to return. Default is 10.
+                distance_type:
+                  type: string
+                  description: |
+                    The distance metric to use for search. L2, Cosine, Dot and Hamming are supported. Default is L2.
+                bypass_vector_index:
+                  type: boolean
+                  description: |
+                    Whether to bypass vector index. Optional.
+                filter:
+                  type: string
+                  description: |
+                    A filter expression that specifies the rows to query. Optional.
+                columns:
+                  type: array
+                  items:
+                    type: string
+                  description: |
+                    The columns to return. Optional.
+                nprobe:
+                  type: integer
+                  description: |
+                    The number of probes to use for search. Optional.
+                refine_factor:
+                  type: integer
+                  description: |
+                    The refine factor to use for search. Optional.
+                  default: null
+                fast_search:
+                  type: boolean
+                  description: |
+                    Whether to use fast search. Optional.
+                  default: false
+              required:
+                - vector
+
+      responses:
+        "200":
+          description: top k results if query is successfully executed
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  results:
+                    type: array
+                    items:
+                      type: object
+                      properties:
+                        id:
+                          type: integer
+                        selected_col_1_to_return:
+                          type: col_1_type
+                        selected_col_n_to_return:
+                          type: col_n_type
+                        _distance:
+                          type: float
+
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+
+  /v1/table/{name}/insert/:
+    post:
+      description: Insert new data to the Table.
+      tags:
+        - Data
+      operationId: insertData
+      summary: Insert new data.
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        $ref: "#/components/requestBodies/arrow_stream_buffer"
+      responses:
+        "200":
+          description: Insert successful
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+  /v1/table/{name}/merge_insert/:
+    post:
+      description: Create a "merge insert" operation
+        This operation can add rows, update rows, and remove rows all in a single
+        transaction. See python method `lancedb.table.Table.merge_insert` for examples.
+      tags:
+        - Data
+      summary: Merge Insert
+      operationId: mergeInsert
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+        - name: on
+          in: query
+          description: |
+            The column to use as the primary key for the merge operation.
+          required: true
+          schema:
+            type: string
+        - name: when_matched_update_all
+          in: query
+          description: |
+            Rows that exist in both the source table (new data) and
+            the target table (old data) will be updated, replacing
+            the old row with the corresponding matching row.
+          required: false
+          schema:
+            type: boolean
+        - name: when_matched_update_all_filt
+          in: query
+          description: |
+            If present then only rows that satisfy the filter expression will
+            be updated
+          required: false
+          schema:
+            type: string
+        - name: when_not_matched_insert_all
+          in: query
+          description: |
+            Rows that exist only in the source table (new data) will be
+            inserted into the target table (old data).
+          required: false
+          schema:
+            type: boolean
+        - name: when_not_matched_by_source_delete
+          in: query
+          description: |
+            Rows that exist only in the target table (old data) will be
+            deleted. An optional condition (`when_not_matched_by_source_delete_filt`)
+            can be provided to limit what data is deleted.
+          required: false
+          schema:
+            type: boolean
+        - name: when_not_matched_by_source_delete_filt
+          in: query
+          description: |
+            The filter expression that specifies the rows to delete.
+          required: false
+          schema:
+            type: string
+      requestBody:
+        $ref: "#/components/requestBodies/arrow_stream_buffer"
+      responses:
+        "200":
+          description: Merge Insert successful
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+  /v1/table/{name}/delete/:
+    post:
+      description: Delete rows from a table.
+      tags:
+        - Data
+      summary: Delete rows from a table
+      operationId: deleteData
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              properties:
+                predicate:
+                  type: string
+                  description: |
+                    A filter expression that specifies the rows to delete.
+      responses:
+        "200":
+          description: Delete successful
+        "401":
+          $ref: "#/components/responses/unauthorized"
+  /v1/table/{name}/drop/:
+    post:
+      description: Drop a table
+      tags:
+        - Tables
+      summary: Drop a table
+      operationId: dropTable
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        $ref: "#/components/requestBodies/arrow_stream_buffer"
+      responses:
+        "200":
+          description: Drop successful
+        "401":
+          $ref: "#/components/responses/unauthorized"
+
+  /v1/table/{name}/describe/:
+    post:
+      description: Describe a table and return Table Information.
+      tags:
+        - Tables
+      summary: Describe a table
+      operationId: describeTable
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      responses:
+        "200":
+          description: Table information
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  table:
+                    type: string
+                  version:
+                    type: integer
+                  schema:
+                    type: string
+                  stats:
+                    type: object
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+
+  /v1/table/{name}/index/list/:
+    post:
+      description: List indexes of a table
+      tags:
+        - Tables
+      summary: List indexes of a table
+      operationId: listIndexes
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      responses:
+        "200":
+          description: Available list of indexes on the table.
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  indexes:
+                    type: array
+                    items:
+                      type: object
+                      properties:
+                        columns:
+                          type: array
+                          items:
+                            type: string
+                        index_name:
+                          type: string
+                        index_uuid:
+                          type: string
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+  /v1/table/{name}/create_index/:
+    post:
+      description: Create vector index on a Table
+      tags:
+        - Tables
+      summary: Create vector index on a Table
+      operationId: createIndex
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              properties:
+                column:
+                  type: string
+                metric_type:
+                  type: string
+                  nullable: false
+                  description: |
+                    The metric type to use for the index. L2, Cosine, Dot are supported.
+                index_type:
+                  type: string
+      responses:
+        "200":
+          description: Index successfully created
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $ref: "#/components/responses/unauthorized"
+        "404":
+          $ref: "#/components/responses/not_found"
+  /v1/table/{name}/create_scalar_index/:
+    post:
+      description: Create a scalar index on a table
+      tags:
+        - Tables
+      summary: Create a scalar index on a table
+      operationId: createScalarIndex
+      parameters:
+        - $ref: "#/components/parameters/table_name"
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              type: object
+              properties:
+                column:
+                  type: string
+                index_type:
+                  type: string
+                  required: false
+      responses:
+        "200":
+          description: Scalar Index successfully created
+        "400":
+          $ref: "#/components/responses/invalid_request"
+        "401":
+          $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/package-lock.json

@@ -0,0 +1,135 @@
+{
+  "name": "lancedb-docs-test",
+  "version": "1.0.0",
+  "lockfileVersion": 3,
+  "requires": true,
+  "packages": {
+    "": {
+      "name": "lancedb-docs-test",
+      "version": "1.0.0",
+      "license": "Apache 2",
+      "dependencies": {
+        "apache-arrow": "file:../node/node_modules/apache-arrow",
+        "vectordb": "file:../node"
+      },
+      "devDependencies": {
+        "@types/node": "^20.11.8",
+        "typescript": "^5.3.3"
+      }
+    },
+    "../node": {
+      "name": "vectordb",
+      "version": "0.12.0",
+      "cpu": [
+        "x64",
+        "arm64"
+      ],
+      "license": "Apache-2.0",
+      "os": [
+        "darwin",
+        "linux",
+        "win32"
+      ],
+      "dependencies": {
+        "@neon-rs/load": "^0.0.74",
+        "axios": "^1.4.0"
+      },
+      "devDependencies": {
+        "@neon-rs/cli": "^0.0.160",
+        "@types/chai": "^4.3.4",
+        "@types/chai-as-promised": "^7.1.5",
+        "@types/mocha": "^10.0.1",
+        "@types/node": "^18.16.2",
+        "@types/sinon": "^10.0.15",
+        "@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",
+        "eslint": "^8.39.0",
+        "eslint-config-standard-with-typescript": "^34.0.1",
+        "eslint-plugin-import": "^2.26.0",
+        "eslint-plugin-n": "^15.7.0",
+        "eslint-plugin-promise": "^6.1.1",
+        "mocha": "^10.2.0",
+        "openai": "^4.24.1",
+        "sinon": "^15.1.0",
+        "temp": "^0.9.4",
+        "ts-node": "^10.9.1",
+        "ts-node-dev": "^2.0.0",
+        "typedoc": "^0.24.7",
+        "typedoc-plugin-markdown": "^3.15.3",
+        "typescript": "^5.1.0",
+        "uuid": "^9.0.0"
+      },
+      "optionalDependencies": {
+        "@lancedb/vectordb-darwin-arm64": "0.12.0",
+        "@lancedb/vectordb-darwin-x64": "0.12.0",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.12.0",
+        "@lancedb/vectordb-linux-x64-gnu": "0.12.0",
+        "@lancedb/vectordb-win32-x64-msvc": "0.12.0"
+      },
+      "peerDependencies": {
+        "@apache-arrow/ts": "^14.0.2",
+        "apache-arrow": "^14.0.2"
+      }
+    },
+    "../node/node_modules/apache-arrow": {
+      "version": "14.0.2",
+      "license": "Apache-2.0",
+      "dependencies": {
+        "@types/command-line-args": "5.2.0",
+        "@types/command-line-usage": "5.0.2",
+        "@types/node": "20.3.0",
+        "@types/pad-left": "2.1.1",
+        "command-line-args": "5.2.1",
+        "command-line-usage": "7.0.1",
+        "flatbuffers": "23.5.26",
+        "json-bignum": "^0.0.3",
+        "pad-left": "^2.1.0",
+        "tslib": "^2.5.3"
+      },
+      "bin": {
+        "arrow2csv": "bin/arrow2csv.js"
+      }
+    },
+    "node_modules/@types/node": {
+      "version": "20.11.8",
+      "resolved": "https://registry.npmjs.org/@types/node/-/node-20.11.8.tgz",
+      "integrity": "sha512-i7omyekpPTNdv4Jb/Rgqg0RU8YqLcNsI12quKSDkRXNfx7Wxdm6HhK1awT3xTgEkgxPn3bvnSpiEAc7a7Lpyow==",
+      "dev": true,
+      "dependencies": {
+        "undici-types": "~5.26.4"
+      }
+    },
+    "node_modules/apache-arrow": {
+      "resolved": "../node/node_modules/apache-arrow",
+      "link": true
+    },
+    "node_modules/typescript": {
+      "version": "5.3.3",
+      "resolved": "https://registry.npmjs.org/typescript/-/typescript-5.3.3.tgz",
+      "integrity": "sha512-pXWcraxM0uxAS+tN0AG/BF2TyqmHO014Z070UsJ+pFvYuRSq8KH8DmWpnbXe0pEPDHXZV3FcAbJkijJ5oNEnWw==",
+      "dev": true,
+      "bin": {
+        "tsc": "bin/tsc",
+        "tsserver": "bin/tsserver"
+      },
+      "engines": {
+        "node": ">=14.17"
+      }
+    },
+    "node_modules/undici-types": {
+      "version": "5.26.5",
+      "resolved": "https://registry.npmjs.org/undici-types/-/undici-types-5.26.5.tgz",
+      "integrity": "sha512-JlCMO+ehdEIKqlFxk6IfVoAUVmgz7cU7zD/h9XZ0qzeosSHmUJVOzSQvvYSYWXkFXC+IfLKSIffhv0sVZup6pA==",
+      "dev": true
+    },
+    "node_modules/vectordb": {
+      "resolved": "../node",
+      "link": true
+    }
+  }
+}

docs/package.json

@@ -0,0 +1,20 @@
+{
+  "name": "lancedb-docs-test",
+  "version": "1.0.0",
+  "description": "auto-generated tests from doc",
+  "author": "dev@lancedb.com",
+  "license": "Apache 2",
+  "dependencies": {
+    "apache-arrow": "file:../node/node_modules/apache-arrow",
+    "vectordb": "file:../node"
+  },
+  "scripts": {
+    "build": "tsc -b && cd ../node && npm run build-release",
+    "example": "npm run build && node",
+    "test": "npm run build && ls dist/*.js | xargs -n 1 node"
+  },
+  "devDependencies": {
+    "@types/node": "^20.11.8",
+    "typescript": "^5.3.3"
+  }
+}

docs/requirements.txt

@@ -1,5 +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

@@ -7,64 +7,87 @@ for brute-force scanning of the entire vector space.
 A vector index is faster but less accurate than exhaustive search (kNN or flat search).
 LanceDB provides many parameters to fine-tune the index's size, the speed of queries, and the accuracy of results.
 
-Currently, LanceDB does *not* automatically create the ANN index.
+## Disk-based Index
-LanceDB has optimized code for kNN as well. For many use-cases, datasets under 100K vectors won't require index creation at all.
-If you can live with <100ms latency, skipping index creation is a simpler workflow while guaranteeing 100% recall.
 
-In the future we will look to automatically create and configure the ANN index as data comes in.
+Lance provides an `IVF_PQ` disk-based index. It uses **Inverted File Index (IVF)** to first divide
-
+the dataset into `N` partitions, and then applies **Product Quantization** to compress vectors in each partition.
-## Types of Index
+See the [indexing](concepts/index_ivfpq.md) concepts guide for more information on how this works.
-
-Lance can support multiple index types, the most widely used one is `IVF_PQ`.
-
-* `IVF_PQ`: use **Inverted File Index (IVF)** to first divide the dataset into `N` partitions,
-   and then use **Product Quantization** to compress vectors in each partition.
-* `DiskANN` (**Experimental**): organize the vector as a on-disk graph, where the vertices approximately
-   represent the nearest neighbors of each vector.
 
 ## Creating an IVF_PQ Index
 
 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.
 
         ```python
-     import lancedb
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb"
-     import numpy as np
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-numpy"
-     uri = "data/sample-lancedb"
+        --8<-- "python/python/tests/docs/test_guide_index.py:create_ann_index"
-     db = lancedb.connect(uri)
+        ```
+    === "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
+        ```python
-     data = [{"vector": row, "item": f"item {i}"}
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb"
-        for i, row in enumerate(np.random.random((10_000, 1536)).astype('float32'))]
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-numpy"
-
+        --8<-- "python/python/tests/docs/test_guide_index.py:import-lancedb-ivfpq"
-     # Add the vectors to a table
+        --8<-- "python/python/tests/docs/test_guide_index.py:create_ann_index_async"
-     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)
         ```
 
-=== "Javascript"
+=== "TypeScript"
-     ```javascript
-     const vectordb = require('vectordb')
-     const db = await vectordb.connect('data/sample-lancedb')
 
-     let data = []
+    === "@lancedb/lancedb"
-     for (let i = 0; i < 10_000; i++) {
+
-         data.push({vector: Array(1536).fill(i), id: `${i}`, content: "", longId: `${i}`},)
+        Creating indexes is done via the [lancedb.Table.createIndex](../js/classes/Table.md/#createIndex) method.
-     }
+
-     const table = await db.createTable('my_vectors', data)
+        ```typescript
-     await table.createIndex({ type: 'ivf_pq', column: 'vector', num_partitions: 256, num_sub_vectors: 96 })
+        --8<--- "nodejs/examples/ann_indexes.test.ts:import"
+
+        --8<-- "nodejs/examples/ann_indexes.test.ts:ingest"
         ```
 
-- **metric** (default: "L2"): The distance metric to use. By default it uses euclidean distance "`L2`".
+    === "vectordb (deprecated)"
+
+        Creating indexes is done via the [lancedb.Table.createIndex](../javascript/interfaces/Table.md/#createIndex) method.
+
+        ```typescript
+        --8<--- "docs/src/ann_indexes.ts:import"
+
+        --8<-- "docs/src/ann_indexes.ts:ingest"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/ivf_pq.rs:create_index"
+    ```
+
+    IVF_PQ index parameters are more fully defined in the [crate docs](https://docs.rs/lancedb/latest/lancedb/index/vector/struct.IvfPqIndexBuilder.html).
+
+The following IVF_PQ paramters can be specified:
+
+- **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** (default: 256): The number of partitions of the index.
+- **num_partitions**: The number of partitions in the index. The default is the square root
-- **num_sub_vectors** (default: 96): The number of sub-vectors (M) that will be created during Product Quantization (PQ).
+  of the number of rows.
-For D dimensional vector, it will be divided into `M` of `D/M` sub-vectors, each of which is presented by
+
-a single PQ code.
+!!! note
+
+    In the synchronous python SDK and node's `vectordb` the default is 256. This default has
+    changed in the asynchronous python SDK and node's `lancedb`.
+
+- **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
+
+    In the synchronous python SDK and node's `vectordb` the default is currently 96. This default has
+    changed in the asynchronous python SDK and node's `lancedb`.
 
 <figure markdown>
   ![IVF PQ](./assets/ivf_pq.png)
@@ -78,7 +101,7 @@ Using GPU for index creation requires [PyTorch>2.0](https://pytorch.org/) being
 
 You can specify the GPU device to train IVF partitions via
 
-- **accelerator**: Specify to ``cuda`` or ``mps`` (on Apple Silicon) to enable GPU training.
+- **accelerator**: Specify to `cuda` or `mps` (on Apple Silicon) to enable GPU training.
 
 === "Linux"
 
@@ -92,7 +115,7 @@ You can specify the GPU device to train IVF partitions via
     )
     ```
 
-=== "Macos"
+=== "MacOS"
 
     <!-- skip-test -->
     ```python
@@ -103,13 +126,14 @@ 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.
     
-Trouble shootings:
+Troubleshooting:
 
-If you see ``AssertionError: Torch not compiled with CUDA enabled``, you need to [install
+If you see `AssertionError: Torch not compiled with CUDA enabled`, you need to [install
 PyTorch with CUDA support](https://pytorch.org/get-started/locally/).
 
-
 ## Querying an ANN Index
 
 Querying vector indexes is done via the [search](https://lancedb.github.io/lancedb/python/#lancedb.table.LanceTable.search) function.
@@ -118,57 +142,86 @@ 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/>
+  Most of the time, setting nprobes to cover 5-15% 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/>
+    - _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/>
-  Note: nprobes is only applicable if an ANN index is present. If specified on a table without an ANN index, it is ignored.
+
 - **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/>
+    - _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: refine_factor is only applicable if an ANN index is present. If specified on a table without an ANN index, it is ignored.
+!!! 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))) \
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search"
-         .limit(2) \
-         .nprobes(20) \
-         .refine_factor(10) \
-         .to_pandas()
         ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_guide_index.py:vector_search_async"
         ```
+
+    ```text
                                               vector       item       _distance
     0  [0.44949695, 0.8444449, 0.06281311, 0.23338133...  item 1141  103.575333
     1  [0.48587373, 0.269207, 0.15095535, 0.65531915,...  item 3953  108.393867
     ```
 
-=== "Javascript"
+=== "TypeScript"
-     ```javascript
+
-     const results_1 = await table
+    === "@lancedb/lancedb"
-         .search(Array(1536).fill(1.2))
+
-         .limit(2)
+        ```typescript
-         .nprobes(20)
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search1"
-         .refineFactor(10)
-         .execute()
         ```
 
-The search will return the data requested in addition to the distance of each item.
+    === "vectordb (deprecated)"
 
+        ```typescript
+        --8<-- "docs/src/ann_indexes.ts:search1"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/ivf_pq.rs:search1"
+    ```
+
+    Vector search options are more fully defined in the [crate docs](https://docs.rs/lancedb/latest/lancedb/query/struct.Query.html#method.nearest_to).
+
+The search will return the data requested in addition to the distance of each item.
 
 ### Filtering (where clause)
 
 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()
+        --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"
         ```
 
-=== "Javascript"
+=== "TypeScript"
+
+    === "@lancedb/lancedb"
+
+        ```typescript
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search2"
+        ```
+
+    === "vectordb (deprecated)"
+
         ```javascript
-     const results_2 = await table
+        --8<-- "docs/src/ann_indexes.ts:search2"
-         .search(Array(1536).fill(1.2))
-         .where("id != '1141'")
-         .limit(2)
-         .execute()
         ```
 
 ### Projections (select clause)
@@ -176,34 +229,56 @@ You can further filter the elements returned by a search using a where clause.
 You can select the columns returned by the query using a select clause.
 
 === "Python"
+
+    === "Sync API"
+
         ```python
-     tbl.search(np.random.random((1536))).select(["vector"]).to_pandas()
+        --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
     0  [0.30928212, 0.022668175, 0.1756372, 0.4911822...  93.971092
     1  [0.2525465, 0.01723831, 0.261568, 0.002007689,...  95.173485
     ...
     ```
 
-=== "Javascript"
+=== "TypeScript"
-     ```javascript
+
-     const results_3 = await table
+    === "@lancedb/lancedb"
-         .search(Array(1536).fill(1.2))
+
-         .select(["id"])
+        ```typescript
-         .limit(2)
+        --8<-- "nodejs/examples/ann_indexes.test.ts:search3"
-         .execute()
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/ann_indexes.ts:search3"
         ```
 
 ## FAQ
 
+### Why do I need to manually create an index?
+
+Currently, LanceDB does _not_ automatically create the ANN index.
+LanceDB is well-optimized for kNN (exhaustive search) via a disk-based index. For many use-cases,
+datasets of the order of ~100K vectors don't require index creation. If you can live with up to
+100ms latency, skipping index creation is a simpler workflow while guaranteeing 100% recall.
+
 ### When is it necessary to create an ANN vector index?
 
-`LanceDB` has manually-tuned SIMD code for computing vector distances.
+`LanceDB` comes out-of-the-box with highly optimized SIMD code for computing vector similarity.
-In our benchmarks, computing 100K pairs of 1K dimension vectors takes **less than 20ms**.
+In our benchmarks, computing distances for 100K pairs of 1K dimension vectors takes **less than 20ms**.
-For small datasets (< 100K rows) or applications that can accept 100ms latency, vector indices are usually not necessary.
+We observe that for small datasets (~100K rows) or for applications that can accept 100ms latency,
+vector indices are usually not necessary.
 
-For large-scale or higher dimension vectors, it is beneficial to create vector index.
+For large-scale or higher dimension vectors, it can beneficial to create vector index for performance.
 
 ### How big is my index, and how many memory will it take?
 
@@ -218,7 +293,15 @@ Product quantization can lead to approximately `16 * sizeof(float32) / 1 = 64` t
 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.
 
-`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
+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.
-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

@@ -0,0 +1,54 @@
+// --8<-- [start:import]
+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");
+
+  let data = [];
+  for (let i = 0; i < 10_000; i++) {
+    data.push({
+      vector: Array(1536).fill(i),
+      id: `${i}`,
+      content: "",
+      longId: `${i}`,
+    });
+  }
+  const table = await db.createTable("my_vectors", data);
+  await table.createIndex({
+    type: "ivf_pq",
+    column: "vector",
+    num_partitions: 16,
+    num_sub_vectors: 48,
+  });
+  // --8<-- [end:ingest]
+
+  // --8<-- [start:search1]
+  const results_1 = await table
+    .search(Array(1536).fill(1.2))
+    .limit(2)
+    .nprobes(20)
+    .refineFactor(10)
+    .execute();
+  // --8<-- [end:search1]
+
+  // --8<-- [start:search2]
+  const results_2 = await table
+    .search(Array(1536).fill(1.2))
+    .where("id != '1141'")
+    .limit(2)
+    .execute();
+  // --8<-- [end:search2]
+
+  // --8<-- [start:search3]
+  const results_3 = await table
+    .search(Array(1536).fill(1.2))
+    .select(["id"])
+    .limit(2)
+    .execute();
+  // --8<-- [end:search3]
+
+  console.log("ann_indexes.ts: done");
+})();

docs/src/api_reference.md

@@ -0,0 +1,8 @@
+# API Reference
+
+The API reference for the LanceDB client SDKs are available at the following locations:
+
+- [Python](python/python.md)
+- [JavaScript (legacy vectordb package)](javascript/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,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"/><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

@@ -3,7 +3,7 @@
 !!! info "LanceDB can be run in a number of ways:"
 
     * Embedded within an existing backend (like your Django, Flask, Node.js or FastAPI application)
-    * Connected to directly from a client application like a Jupyter notebook for analytical workloads
+    * Directly from a client application like a Jupyter notebook for analytical workloads
     * Deployed as a remote serverless database
 
 ![](assets/lancedb_embedded_explanation.png)
@@ -11,207 +11,47 @@
 ## Installation
 
 === "Python"
+
       ```shell
       pip install lancedb
       ```
 
-=== "Javascript"
+=== "Typescript[^1]"
+    === "@lancedb/lancedb"
+
+        ```shell
+        npm install @lancedb/lancedb
+        ```
+        !!! note "Bundling `@lancedb/lancedb` apps with Webpack"
+
+            Since LanceDB contains a prebuilt Node binary, you must configure `next.config.js` to exclude it from webpack. This is required for both using Next.js and deploying a LanceDB app on Vercel.
+
+            ```javascript
+            /** @type {import('next').NextConfig} */
+            module.exports = ({
+            webpack(config) {
+                config.externals.push({ '@lancedb/lancedb': '@lancedb/lancedb' })
+                return config;
+            }
+            })
+            ```
+
+        !!! 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
         npm install vectordb
         ```
-
-## How to connect to a database
-
-=== "Python"
-      ```python
-      import lancedb
-      uri = "data/sample-lancedb"
-      db = lancedb.connect(uri)
-      ```
-
-      LanceDB will create the directory if it doesn't exist (including parent directories).
-
-      If you need a reminder of the uri, use the `db.uri` property.
-
-=== "Javascript"
-      ```javascript
-      const lancedb = require("vectordb");
-
-      const uri = "data/sample-lancedb";
-      const db = await lancedb.connect(uri);
-      ```
-      
-      LanceDB will create the directory if it doesn't exist (including parent directories).
-
-      If you need a reminder of the uri, you can call `db.uri()`.
-
-## How to create a table
-
-=== "Python"
-    ```python
-    tbl = 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}])
-    ```
-
-    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:
-    ```python
-    import pandas as pd
-    df = pd.DataFrame([{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-                       {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}])
-    tbl = db.create_table("table_from_df", data=df)
-    ```
-
-=== "Javascript"
-    ```javascript
-    const tb = await db.createTable(
-        "myTable",
-        [{"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
-         {"vector": [5.9, 26.5], "item": "bar", "price": 20.0}]
-    )
-    ```
-
-    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 `createTable` function.
-
-
-!!! info "Under the hood, LanceDB is converting the input data into an Apache Arrow table and persisting it to disk in [Lance format](https://www.github.com/lancedb/lance)."
-
-### Creating an empty table
-
-Sometimes you may not have the data to insert into the table at creation time.
-In this case, you can create an empty table and specify the schema.
-
-=== "Python"
-      ```python
-      import pyarrow as pa
-      schema = pa.schema([pa.field("vector", pa.list_(pa.float32(), list_size=2))])
-      tbl = db.create_table("empty_table", schema=schema)
-      ```
-
-## How to open an existing table
-
-Once created, you can open a table using the following code:
-
-=== "Python"
-      ```python
-      tbl = db.open_table("my_table")
-      ```
-
-      If you forget the name of your table, you can always get a listing of all table names:
-
-      ```python
-      print(db.table_names())
-      ```
-
-=== "Javascript"
-      ```javascript
-      const tbl = await db.openTable("myTable");
-      ```
-
-      If you forget the name of your table, you can always get a listing of all table names:
-
-      ```javascript
-      console.log(await db.tableNames());
-      ```
-
-## How to add data to a table
-
-After a table has been created, you can always add more data to it using
-
-=== "Python"
-      ```python
-
-      # Option 1: Add a list of dicts to a table
-      data = [{"vector": [1.3, 1.4], "item": "fizz", "price": 100.0},
-              {"vector": [9.5, 56.2], "item": "buzz", "price": 200.0}]
-      tbl.add(data)
-
-      # Option 2: Add a pandas DataFrame to a table
-      df = pd.DataFrame(data)
-      tbl.add(data)
-      ```
-
-=== "Javascript"
-      ```javascript
-      await tbl.add([{vector: [1.3, 1.4], item: "fizz", price: 100.0},
-                     {vector: [9.5, 56.2], item: "buzz", price: 200.0}])
-      ```
-
-## How to search for (approximate) nearest neighbors
-
-Once you've embedded the query, you can find its nearest neighbors using the following code:
-
-=== "Python"
-      ```python
-      tbl.search([100, 100]).limit(2).to_pandas()
-      ```
-
-      This returns a pandas DataFrame with the results.
-
-=== "Javascript"
-      ```javascript
-      const query = await tbl.search([100, 100]).limit(2).execute();
-      ```
-
-## How to delete rows from a table
-
-Use the `delete()` method on tables to delete rows from a table. To choose
-which rows to delete, provide a filter that matches on the metadata columns.
-This can delete any number of rows that match the filter.
-
-=== "Python"
-      ```python
-      tbl.delete('item = "fizz"')
-      ```
-
-=== "Javascript"
-      ```javascript
-      await tbl.delete('item = "fizz"')
-      ```
-
-The deletion predicate is a SQL expression that supports the same expressions
-as the `where()` clause on a search. They can be as simple or complex as needed.
-To see what expressions are supported, see the [SQL filters](sql.md) section.
-
-
-=== "Python"
-
-      Read more: [lancedb.table.Table.delete][]
-
-=== "Javascript"
-
-      Read more: [vectordb.Table.delete](javascript/interfaces/Table.md#delete)
-
-## How to remove a table
-
-Use the `drop_table()` method on the database to remove a table.
-
-=== "Python"
-      ```python
-      db.drop_table("my_table")
-      ```
-
-      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,
-      you can pass in `ignore_missing=True`.
-
-=== "JavaScript"
-      ```javascript
-      await db.dropTable('myTable')
-      ```
-
-      This permanently removes the table and is not recoverable, unlike deleting rows.
-      If the table does not exist an exception is raised. 
-
         !!! note "Bundling `vectordb` apps with Webpack"
 
-    If you're using the `vectordb` module in JavaScript, since LanceDB contains a prebuilt Node binary, you must configure `next.config.js` to exclude it from webpack. This is required for both using Next.js and deploying a LanceDB app on Vercel.
+            Since LanceDB contains a prebuilt Node binary, you must configure `next.config.js` to exclude it from webpack. This is required for both using Next.js and deploying a LanceDB app on Vercel.
 
             ```javascript
             /** @type {import('next').NextConfig} */
@@ -223,8 +63,593 @@ Use the `drop_table()` method on the database to remove a table.
             })
             ```
 
+        !!! 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
+    cargo add lancedb
+    ```
+
+    !!! info "To use the lancedb create, you first need to install protobuf."
+
+    === "macOS"
+
+        ```shell
+        brew install protobuf
+        ```
+
+    === "Ubuntu/Debian"
+
+        ```shell
+        sudo apt install -y protobuf-compiler libssl-dev
+        ```
+
+    !!! info "Please also make sure you're using the same version of Arrow as in the [lancedb crate](https://github.com/lancedb/lancedb/blob/main/Cargo.toml)"
+
+### Preview releases
+
+Stable releases are created about every 2 weeks. For the latest features and bug
+fixes, you can install the preview release. These releases receive the same
+level of testing as stable releases, but are not guaranteed to be available for
+more than 6 months after they are released. Once your application is stable, we
+recommend switching to stable releases.
+
+=== "Python"
+
+      ```shell
+      pip install --pre --extra-index-url https://pypi.fury.io/lancedb/ lancedb
+      ```
+
+=== "Typescript[^1]"
+
+    === "@lancedb/lancedb"
+
+        ```shell
+        npm install @lancedb/lancedb@preview
+        ```
+    === "vectordb (deprecated)"
+
+        ```shell
+        npm install vectordb@preview
+        ```
+
+=== "Rust"
+
+    We don't push preview releases to crates.io, but you can referent the tag
+    in GitHub within your Cargo dependencies:
+
+    ```toml
+    [dependencies]
+    lancedb = { git = "https://github.com/lancedb/lancedb.git", tag = "vX.Y.Z-beta.N" }
+    ```
+
+## 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:set_uri"
+        --8<-- "python/python/tests/docs/test_basic.py:connect"
+        ```
+    === "Async API"
+
+        ```python
+        --8<-- "python/python/tests/docs/test_basic.py:imports"
+
+        --8<-- "python/python/tests/docs/test_basic.py:set_uri"
+        --8<-- "python/python/tests/docs/test_basic.py:connect_async"
+        ```
+
+=== "Typescript[^1]"
+
+    === "@lancedb/lancedb"
+
+        ```typescript
+        import * as lancedb from "@lancedb/lancedb";
+        import * as arrow from "apache-arrow";
+
+        --8<-- "nodejs/examples/basic.test.ts:connect"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:open_db"
+        ```
+
+=== "Rust"
+
+    ```rust
+    #[tokio::main]
+    async fn main() -> Result<()> {
+        --8<-- "rust/lancedb/examples/simple.rs:connect"
+    }
+    ```
+
+    !!! info "See [examples/simple.rs](https://github.com/lancedb/lancedb/tree/main/rust/lancedb/examples/simple.rs) for a full working example."
+
+LanceDB will create the directory if it doesn't exist (including parent directories).
+
+If you need a reminder of the uri, you can call `db.uri()`.
+
+## Create a table
+
+### Create a table from initial data
+
+If you have data to insert into the table at creation time, you can simultaneously create a
+table and insert the data into it. The schema of the data will be used as the schema of the
+table.
+
+=== "Python"
+
+    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.
+
+    === "Sync API"
+
+        ```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.test.ts:create_table"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:create_table"
+        ```
+
+        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 `createTable` function.
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:create_table"
+    ```
+
+    If the table already exists, LanceDB will raise an error by default.  See
+    [the mode option](https://docs.rs/lancedb/latest/lancedb/connection/struct.CreateTableBuilder.html#method.mode)
+    for details on how to overwrite (or open) existing tables instead.
+
+    !!! Providing table records in Rust
+
+        The Rust SDK currently expects data to be provided as an Arrow
+        [RecordBatchReader](https://docs.rs/arrow-array/latest/arrow_array/trait.RecordBatchReader.html)
+        Support for additional formats (such as serde or polars) is on the roadmap.
+
+!!! info "Under the hood, LanceDB reads in the Apache Arrow data and persists it to disk using the [Lance format](https://www.github.com/lancedb/lance)."
+
+!!! info "Automatic embedding generation with Embedding API"
+    When working with embedding models, it is recommended to use the LanceDB embedding API to automatically create vector representation of the data and queries in the background. See the [quickstart example](#using-the-embedding-api) or the embedding API [guide](./embeddings/)
+
+### Create an empty table
+
+Sometimes you may not have the data to insert into the table at creation time.
+In this case, you can create an empty table and specify the schema, so that you can add
+data to the table at a later time (as long as it conforms to the schema). This is
+similar to a `CREATE TABLE` statement in SQL.
+
+=== "Python"
+
+    === "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).
+
+=== "Typescript[^1]"
+
+    === "@lancedb/lancedb"
+
+        ```typescript
+        --8<-- "nodejs/examples/basic.test.ts:create_empty_table"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:create_empty_table"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:create_empty_table"
+    ```
+
+## Open an existing table
+
+Once created, you can open a table as follows:
+
+=== "Python"
+
+    === "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.test.ts:open_table"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        const tbl = await db.openTable("myTable");
+        ```
+
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:open_existing_tbl"
+    ```
+
+If you forget the name of your table, you can always get a listing of all table names:
+
+=== "Python"
+
+    === "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.test.ts:table_names"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        console.log(await db.tableNames());
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:list_names"
+    ```
+
+## Add data to a table
+
+After a table has been created, you can always add more data to it as follows:
+
+=== "Python"
+
+    === "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.test.ts:add_data"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:add"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:add"
+    ```
+
+## Search for nearest neighbors
+
+Once you've embedded the query, you can find its nearest neighbors as follows:
+
+=== "Python"
+
+    === "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.
+
+=== "Typescript[^1]"
+    === "@lancedb/lancedb"
+
+        ```typescript
+        --8<-- "nodejs/examples/basic.test.ts:vector_search"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:search"
+        ```
+
+=== "Rust"
+
+    ```rust
+    use futures::TryStreamExt;
+
+    --8<-- "rust/lancedb/examples/simple.rs:search"
+    ```
+
+    !!! Query vectors in Rust
+        Rust does not yet support automatic execution of embedding functions.  You will need to
+        calculate embeddings yourself.  Support for this is on the roadmap and can be tracked at
+        https://github.com/lancedb/lancedb/issues/994
+
+        Query vectors can be provided as Arrow arrays or a Vec/slice of Rust floats.
+        Support for additional formats (e.g. `polars::series::Series`) is on the roadmap.
+
+By default, LanceDB runs a brute-force scan over dataset to find the K nearest neighbours (KNN).
+For tables with more than 50K vectors, creating an ANN index is recommended to speed up search performance.
+LanceDB allows you to create an ANN index on a table as follows:
+
+=== "Python"
+
+    === "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.test.ts:create_index"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```{.typescript .ignore}
+        --8<-- "docs/src/basic_legacy.ts:create_index"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:create_index"
+    ```
+
+!!! note "Why do I need to create an index manually?"
+LanceDB does not automatically create the ANN index for two reasons. The first is that it's optimized
+for really fast retrievals via a disk-based index, and the second is that data and query workloads can
+be very diverse, so there's no one-size-fits-all index configuration. LanceDB provides many parameters
+to fine-tune index size, query latency and accuracy. See the section on
+[ANN indexes](ann_indexes.md) for more details.
+
+## Delete rows from a table
+
+Use the `delete()` method on tables to delete rows from a table. To choose
+which rows to delete, provide a filter that matches on the metadata columns.
+This can delete any number of rows that match the filter.
+
+=== "Python"
+
+    === "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.test.ts:delete_rows"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:delete"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:delete"
+    ```
+
+The deletion predicate is a SQL expression that supports the same expressions
+as the `where()` clause (`only_if()` in Rust) on a search. They can be as
+simple or complex as needed. To see what expressions are supported, see the
+[SQL filters](sql.md) section.
+
+=== "Python"
+
+    === "Sync API"
+        Read more: [lancedb.table.Table.delete][]
+    === "Async API"
+        Read more: [lancedb.table.AsyncTable.delete][]
+
+=== "Typescript[^1]"
+
+    === "@lancedb/lancedb"
+
+        Read more: [lancedb.Table.delete](javascript/interfaces/Table.md#delete)
+
+    === "vectordb (deprecated)"
+
+        Read more: [vectordb.Table.delete](javascript/interfaces/Table.md#delete)
+
+=== "Rust"
+
+      Read more: [lancedb::Table::delete](https://docs.rs/lancedb/latest/lancedb/table/struct.Table.html#method.delete)
+
+## Drop a table
+
+Use the `drop_table()` method on the database to remove a table.
+
+=== "Python"
+
+    === "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,
+    you can pass in `ignore_missing=True`.
+
+=== "Typescript[^1]"
+
+    === "@lancedb/lancedb"
+
+        ```typescript
+        --8<-- "nodejs/examples/basic.test.ts:drop_table"
+        ```
+
+    === "vectordb (deprecated)"
+
+        ```typescript
+        --8<-- "docs/src/basic_legacy.ts:drop_table"
+        ```
+
+        This permanently removes the table and is not recoverable, unlike deleting rows.
+        If the table does not exist an exception is raised.
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/simple.rs:drop_table"
+    ```
+
+
+## Using the Embedding API
+You can use the embedding API when working with embedding models. It automatically vectorizes the data at ingestion and query time and comes with built-in integrations with popular embedding models like Openai, Hugging Face, Sentence Transformers, CLIP and more.
+
+=== "Python"
+
+    === "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.test.ts:imports"
+        --8<-- "nodejs/examples/embedding.test.ts:openai_embeddings"
+        ```
+
+=== "Rust"
+
+    ```rust
+    --8<-- "rust/lancedb/examples/openai.rs:imports"
+    --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/index.md).
+
+
 ## What's next
 
 This section covered the very basics of using LanceDB. If you're learning about vector databases for the first time, you may want to read the page on [indexing](concepts/index_ivfpq.md) to get familiar with the concepts.
 
 If you've already worked with other vector databases, you may want to read the [guides](guides/tables.md) to learn how to work with LanceDB in more detail.
+
+[^1]: The `vectordb` package is a legacy package that is  deprecated in favor of `@lancedb/lancedb`.  The `vectordb` package will continue to receive bug fixes and security updates until September 2024.  We recommend all new projects use `@lancedb/lancedb`.  See the [migration guide](migration.md) for more information.

docs/src/basic_legacy.ts

@@ -0,0 +1,126 @@
+// --8<-- [start:import]
+import * as lancedb from "vectordb";
+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";
+
+const example = async () => {
+  fs.rmSync("data/sample-lancedb", { recursive: true, force: true });
+  // --8<-- [start:open_db]
+  const lancedb = require("vectordb");
+  const uri = "data/sample-lancedb";
+  const db = await lancedb.connect(uri);
+  // --8<-- [end:open_db]
+
+  // --8<-- [start:create_table]
+  const tbl = await db.createTable(
+    "myTable",
+    [
+      { vector: [3.1, 4.1], item: "foo", price: 10.0 },
+      { vector: [5.9, 26.5], item: "bar", price: 20.0 },
+    ],
+    { 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) => ({
+    vector: [i, i + 1],
+    item: "fizz",
+    price: i * 0.1,
+  }));
+  await tbl.add(newData);
+  // --8<-- [end:add]
+
+  // --8<-- [start:create_index]
+  await tbl.createIndex({
+    type: "ivf_pq",
+    num_partitions: 2,
+    num_sub_vectors: 2,
+  });
+  // --8<-- [end:create_index]
+
+  // --8<-- [start:create_empty_table]
+  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 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(
+      Array.from(Array(total), (_, i) => ({
+        id: i,
+        vector: Array.from(Array(dim), Math.random),
+      })),
+      { 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]
+
+  // --8<-- [start:delete]
+  await tbl.delete('item = "fizz"');
+  // --8<-- [end:delete]
+
+  // --8<-- [start:drop_table]
+  await db.dropTable("myTable");
+  // --8<-- [end:drop_table]
+};
+
+async function main() {
+  console.log("basic_legacy.ts: start");
+  await example();
+  console.log("basic_legacy.ts: done");
+}
+
+main();

docs/src/cli_config.md

@@ -1,51 +0,0 @@
-
-# CLI & Config
-
-## LanceDB CLI
-Once lanceDB is installed, you can access the CLI using `lancedb` command on the console.
-
-```
-lancedb
-```
-
-This lists out all the various command-line options available. You can get the usage or help for a particular command.
-
-```
-lancedb {command} --help
-```
-
-## LanceDB config
-LanceDB uses a global config file to store certain settings. These settings are configurable using the lanceDB cli.
-To view your config settings, you can use:
-
-```
-lancedb config
-```
-
-These config parameters can be tuned using the cli.
-
-```
-lancedb {config_name} --{argument}
-```
-
-## LanceDB Opt-in Diagnostics
-When enabled, LanceDB will send anonymous events to help us improve LanceDB. These diagnostics are used only for error reporting and no data is collected. Error & stats allow us to automate certain aspects of bug reporting, prioritization of fixes and feature requests.
-These diagnostics are opt-in and can be enabled or disabled using the `lancedb diagnostics` command. These are enabled by default.
-
-### Get usage help
-
-```
-lancedb diagnostics --help
-```
-
-### Disable diagnostics
-
-```
-lancedb diagnostics --disabled
-```
-
-### Enable diagnostics
-
-```
-lancedb diagnostics --enabled
-```

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/rest.md

@@ -0,0 +1 @@
+!!swagger ../../openapi.yml!!

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

@@ -31,7 +31,7 @@ As an example, consider starting with 128-dimensional vector consisting of 32-bi
 
 While PQ helps with reducing the size of the index, IVF primarily addresses search performance. The primary purpose of an inverted file index is to facilitate rapid and effective nearest neighbor search by narrowing down the search space.
 
-In IVF, the PQ vector space is divided into *Voronoi cells*, which are essentially partitions that consist of all the points in the space that are within a threshold distance of the given region's seed point. These seed points are used to create an inverted index that correlates each centroid with a list of vectors in the space, allowing a search to be restricted to just a subset of vectors in the index.
+In IVF, the PQ vector space is divided into *Voronoi cells*, which are essentially partitions that consist of all the points in the space that are within a threshold distance of the given region's seed point. These seed points are initialized by running K-means over the stored vectors. The centroids of K-means turn into the seed points which then each define a region. These regions are then are used to create an inverted index that correlates each centroid with a list of vectors in the space, allowing a search to be restricted to just a subset of vectors in the index.
 
 ![](../assets/ivfpq_ivf_desc.webp)
 
@@ -58,8 +58,10 @@ In Python, the index can be created as follows:
 # 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)
 ```
+!!! 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
@@ -81,24 +83,4 @@ The above query will perform a search on the table `tbl` using the given query v
 * `to_pandas()`: Convert the results to a pandas DataFrame
 
 And there you have it! You now understand what an IVF-PQ index is, and how to create and query it in LanceDB.
-
+To see how to create an IVF-PQ index in LanceDB, take a look at the [ANN indexes](../ann_indexes.md) section.
-
-## FAQ
-
-### When is it necessary to create a vector index?
-
-LanceDB has manually-tuned SIMD code for computing vector distances. In our benchmarks, computing 100K pairs of 1K dimension vectors takes **<20ms**. For small datasets (<100K rows) or applications that can accept up to 100ms latency, vector indices are usually not necessary.
-
-For large-scale or higher dimension vectors, it is beneficial to create vector index.
-
-### How big is my index, and how much memory will it take?
-
-In LanceDB, all vector indices are disk-based, meaning that when responding to a vector query, only the relevant pages from the index file are loaded from disk and cached in memory. Additionally, each sub-vector is usually encoded into 1 byte PQ code.
-
-For example, with 1024-dimension vectors, if we choose `num_sub_vectors = 64`, each sub-vector has `1024 / 64 = 16` float32 numbers. Product quantization can lead to approximately `16 * sizeof(float32) / 1 = 64` times of space reduction.
-
-### How to choose `num_partitions` and `num_sub_vectors` for IVF_PQ index?
-
-`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.
-
-`num_sub_vectors` specifies how many PQ short codes to generate on each vector. 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.

docs/src/embeddings/api.md

@@ -1,211 +0,0 @@
-To use your own custom embedding function, you can follow these 2 simple steps:
-
-1. Create your embedding function by implementing the `EmbeddingFunction` interface
-2. Register your embedding function in the global `EmbeddingFunctionRegistry`.
-
-Let us see how this looks like in action.
-
-![](../assets/embeddings_api.png)
-
-`EmbeddingFunction` and `EmbeddingFunctionRegistry` handle low-level details for serializing schema and model information as metadata. To build a custom embedding function, you don't have to worry about the finer details - simply focus on setting up the model and leave the rest to LanceDB.
-
-## `TextEmbeddingFunction` interface
-
-There is another optional layer of abstraction available: `TextEmbeddingFunction`. You can use this abstraction if your model isn't multi-modal in nature and only needs to operate on text. In such cases, both the source and vector fields will have the same work for vectorization, so you simply just need to setup the model and rest is handled by `TextEmbeddingFunction`. You can read more about the class and its attributes in the class reference.
-
-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
-
-@register("sentence-transformers")
-class SentenceTransformerEmbeddings(TextEmbeddingFunction):
-    name: str = "all-MiniLM-L6-v2"
-    # set more default instance vars like device, etc.
-
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-        self._ndims = None
-    
-    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)
-```
-
-This is a stripped down version of our implementation of `SentenceTransformerEmbeddings` that removes certain optimizations and defaul settings.
-
-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
-
-registry = EmbeddingFunctionRegistry.get_instance()
-stransformer = registry.get("sentence-transformers").create()
-
-class TextModelSchema(LanceModel):
-    vector: Vector(stransformer.ndims) = stransformer.VectorField()
-    text: str = stransformer.SourceField()
-
-tbl = db.create_table("table", schema=TextModelSchema)
-
-tbl.add(pd.DataFrame({"text": ["halo", "world"]}))
-result = tbl.search("world").limit(5)
-```
-
-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.
-
-```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 __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        open_clip = self.safe_import("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
-
-    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 = self.safe_import("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 = self.safe_import("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 = self.safe_import("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 = self.safe_import("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()
-```

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)
+```