Bump version: 0.23.1-beta.2 → 0.23.1

2026-03-28 11:30:39 +00:00 · 2026-01-02 17:38:59 +00:00
65 changed files with 930 additions and 1982 deletions
--- a/.bumpversion.toml
+++ b/.bumpversion.toml
@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.24.0"
+current_version = "0.23.1"
 parse = """(?x)
    (?P<major>0|[1-9]\\d*)\\.
    (?P<minor>0|[1-9]\\d*)\\.
--- a/.github/workflows/codex-update-lance-dependency.yml
+++ b/.github/workflows/codex-update-lance-dependency.yml
@@ -75,13 +75,6 @@ jobs:
          VERSION="${VERSION#v}"
          BRANCH_NAME="codex/update-lance-${VERSION//[^a-zA-Z0-9]/-}"

-          # Use "chore" for beta/rc versions, "feat" for stable releases
-          if [[ "${VERSION}" == *beta* ]] || [[ "${VERSION}" == *rc* ]]; then
-            COMMIT_TYPE="chore"
-          else
-            COMMIT_TYPE="feat"
-          fi
-
          cat <<EOF >/tmp/codex-prompt.txt
          You are running inside the lancedb repository on a GitHub Actions runner. Update the Lance dependency to version ${VERSION} and prepare a pull request for maintainers to review.

@@ -91,10 +84,10 @@ jobs:
          3. After clippy succeeds, run "cargo fmt --all" to format the workspace.
          4. Ensure the repository is clean except for intentional changes. Inspect "git status --short" and "git diff" to confirm the dependency update and any required fixes.
          5. Create and switch to a new branch named "${BRANCH_NAME}" (replace any duplicated hyphens if necessary).
-          6. Stage all relevant files with "git add -A". Commit using the message "${COMMIT_TYPE}: update lance dependency to v${VERSION}".
+          6. Stage all relevant files with "git add -A". Commit using the message "chore: update lance dependency to v${VERSION}".
          7. Push the branch to origin. If the branch already exists, force-push your changes.
          8. env "GH_TOKEN" is available, use "gh" tools for github related operations like creating pull request.
-          9. Create a pull request targeting "main" with title "${COMMIT_TYPE}: update lance dependency to v${VERSION}". First, write the PR body to /tmp/pr-body.md using a heredoc (cat <<'EOF' > /tmp/pr-body.md). The body should summarize the dependency bump, clippy/fmt verification, and link the triggering tag (${TAG}). Then run "gh pr create --body-file /tmp/pr-body.md".
+          9. Create a pull request targeting "main" with title "chore: update lance dependency to v${VERSION}". In the body, summarize the dependency bump, clippy/fmt verification, and link the triggering tag (${TAG}).
          10. After creating the PR, display the PR URL, "git status --short", and a concise summary of the commands run and their results.

          Constraints:
--- a/.github/workflows/rust.yml
+++ b/.github/workflows/rust.yml
@@ -167,13 +167,13 @@ jobs:
      - name: Build
        run: |
          $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
-          cargo build --profile ci --features aws,remote --tests --locked --target ${{ matrix.target }}
+          cargo build --profile ci --features remote --tests --locked --target ${{ matrix.target }}
      - name: Run tests
        # Can only run tests when target matches host
        if: ${{ matrix.target == 'x86_64-pc-windows-msvc' }}
        run: |
          $env:VCPKG_ROOT = $env:VCPKG_INSTALLATION_ROOT
-          cargo test --profile ci --features aws,remote --locked
+          cargo test --profile ci --features remote --locked

  msrv:
    # Check the minimum supported Rust version
--- a/Cargo.lock
+++ b/Cargo.lock
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -15,39 +15,39 @@ categories = ["database-implementations"]
 rust-version = "1.78.0"

 [workspace.dependencies]
-lance = { "version" = "=2.0.0-rc.1", default-features = false, "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-core = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-datagen = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-file = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-io = { "version" = "=2.0.0-rc.1", default-features = false, "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-index = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-linalg = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace-impls = { "version" = "=2.0.0-rc.1", default-features = false, "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-table = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-testing = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-datafusion = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-encoding = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
-lance-arrow = { "version" = "=2.0.0-rc.1", "tag" = "v2.0.0-rc.1", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=1.0.1", default-features = false }
+lance-core = "=1.0.1"
+lance-datagen = "=1.0.1"
+lance-file = "=1.0.1"
+lance-io = { "version" = "=1.0.1", default-features = false }
+lance-index = "=1.0.1"
+lance-linalg = "=1.0.1"
+lance-namespace = "=1.0.1"
+lance-namespace-impls = { "version" = "=1.0.1", default-features = false }
+lance-table = "=1.0.1"
+lance-testing = "=1.0.1"
+lance-datafusion = "=1.0.1"
+lance-encoding = "=1.0.1"
+lance-arrow = "=1.0.1"
 ahash = "0.8"
 # Note that this one does not include pyarrow
-arrow = { version = "57.2", optional = false }
-arrow-array = "57.2"
-arrow-data = "57.2"
-arrow-ipc = "57.2"
-arrow-ord = "57.2"
-arrow-schema = "57.2"
-arrow-select = "57.2"
-arrow-cast = "57.2"
+arrow = { version = "56.2", optional = false }
+arrow-array = "56.2"
+arrow-data = "56.2"
+arrow-ipc = "56.2"
+arrow-ord = "56.2"
+arrow-schema = "56.2"
+arrow-select = "56.2"
+arrow-cast = "56.2"
 async-trait = "0"
-datafusion = { version = "51.0", default-features = false }
-datafusion-catalog = "51.0"
-datafusion-common = { version = "51.0", default-features = false }
-datafusion-execution = "51.0"
-datafusion-expr = "51.0"
-datafusion-physical-plan = "51.0"
+datafusion = { version = "50.1", default-features = false }
+datafusion-catalog = "50.1"
+datafusion-common = { version = "50.1", default-features = false }
+datafusion-execution = "50.1"
+datafusion-expr = "50.1"
+datafusion-physical-plan = "50.1"
 env_logger = "0.11"
-half = { "version" = "2.7.1", default-features = false, features = [
+half = { "version" = "2.6.0", default-features = false, features = [
    "num-traits",
 ] }
 futures = "0"
--- a/ci/run_with_test_connection.sh
+++ b/ci/run_with_test_connection.sh
@@ -16,7 +16,7 @@ check_command_exists() {
 }

 if [[ ! -e ./lancedb ]]; then
-    if [[ x${SOPHON_READ_TOKEN} != "x" ]]; then
+    if [[ -v SOPHON_READ_TOKEN ]]; then
        INPUT="lancedb-linux-x64"
        gh release \
            --repo lancedb/lancedb \
--- a/docs/mkdocs.yml
+++ b/docs/mkdocs.yml
@@ -11,7 +11,7 @@ watch:
 theme:
  name: "material"
  logo: assets/logo.png
-  favicon: assets/favicon.ico
+  favicon: assets/logo.png
  palette:
    # Palette toggle for light mode
    - scheme: lancedb
@@ -32,6 +32,8 @@ theme:
    - content.tooltips
    - toc.follow
    - navigation.top
+    - navigation.tabs
+    - navigation.tabs.sticky
    - navigation.footer
    - navigation.tracking
    - navigation.instant
@@ -113,13 +115,12 @@ markdown_extensions:
      emoji_index: !!python/name:material.extensions.emoji.twemoji
      emoji_generator: !!python/name:material.extensions.emoji.to_svg
  - markdown.extensions.toc:
-      toc_depth: 3
-      permalink: true
-      permalink_title: Anchor link to this section
+      baselevel: 1
+      permalink: ""

 nav:
-  - Documentation:
-      - SDK Reference: index.md
+  - API reference:
+      - Overview: index.md
      - Python: python/python.md
      - Javascript/TypeScript: js/globals.md
      - Java: java/java.md
--- a/docs/src/assets/favicon.ico
+++ b/docs/src/assets/favicon.ico
--- a/docs/src/embeddings/available_embedding_models/multimodal_embedding_functions/voyageai_multimodal_embedding.md
+++ b/docs/src/embeddings/available_embedding_models/multimodal_embedding_functions/voyageai_multimodal_embedding.md
@@ -1,111 +0,0 @@
-# VoyageAI Embeddings : Multimodal
-
-VoyageAI 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://docs.voyageai.com/docs/multimodal-embeddings](https://docs.voyageai.com/docs/multimodal-embeddings)
-
-Supported multimodal models:
-
- `voyage-multimodal-3` - 1024 dimensions (text + images)
- `voyage-multimodal-3.5` - Flexible dimensions (256, 512, 1024 default, 2048). Supports text, images, and video.
-
-### Video Support (voyage-multimodal-3.5)
-
-The `voyage-multimodal-3.5` model supports video input through:
- Video URLs (`.mp4`, `.webm`, `.mov`, `.avi`, `.mkv`, `.m4v`, `.gif`)
- Video file paths
-
-Constraints: Max 20MB video size.
-
-Supported parameters (to be passed in `create` method) are:
-
-| Parameter | Type | Default Value           | Description                               |
-|---|---|-------------------------|-------------------------------------------|
-| `name` | `str` | `"voyage-multimodal-3"` | The model ID of the VoyageAI model to use |
-| `output_dimension` | `int` | `None` | Output dimension for voyage-multimodal-3.5. Valid: 256, 512, 1024, 2048 |
-
-Usage Example:
-
-```python
-import base64
-import os
-from io import BytesIO
-
-import requests
-import lancedb
-from lancedb.pydantic import LanceModel, Vector
-from lancedb.embeddings import get_registry
-import pandas as pd
-
-os.environ['VOYAGE_API_KEY'] = 'YOUR_VOYAGE_API_KEY'
-
-db = lancedb.connect(".lancedb")
-func = get_registry().get("voyageai").create(name="voyage-multimodal-3")
-
-
-def image_to_base64(image_bytes: bytes):
-    buffered = BytesIO(image_bytes)
-    img_str = base64.b64encode(buffered.getvalue())
-    return img_str.decode("utf-8")
-
-
-class Images(LanceModel):
-    label: str
-    image_uri: str = func.SourceField()  # image uri as the source
-    image_bytes: str = func.SourceField()  # image bytes base64 encoded as the source
-    vector: Vector(func.ndims()) = func.VectorField()  # vector column
-    vec_from_bytes: Vector(func.ndims()) = func.VectorField()  # Another vector column
-
-
-if "images" in db.table_names():
-    db.drop_table("images")
-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
-images_bytes = [image_to_base64(requests.get(uri).content) for uri in uris]
-table.add(
-    pd.DataFrame({"label": labels, "image_uri": uris, "image_bytes": images_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", "vec_from_bytes").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(BytesIO(image_bytes))
-actual = table.search(query_image, "vec_from_bytes").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)
-
-```
--- a/docs/src/index.md
+++ b/docs/src/index.md
@@ -1,12 +1,8 @@
-# SDK Reference
+# API Reference

-This site contains the API reference for the client SDKs supported by [LanceDB](https://lancedb.com).
+This page contains the API reference for the SDKs supported by the LanceDB team.

 - [Python](python/python.md)
 - [JavaScript/TypeScript](js/globals.md)
 - [Java](java/java.md)
- [Rust](https://docs.rs/lancedb/latest/lancedb/index.html)
-
-!!! info "LanceDB Documentation"
-
-    If you're looking for the full documentation of LanceDB, visit [docs.lancedb.com](https://docs.lancedb.com).
+- [Rust](https://docs.rs/lancedb/latest/lancedb/index.html)
--- a/docs/src/java/java.md
+++ b/docs/src/java/java.md
@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
 <dependency>
    <groupId>com.lancedb</groupId>
    <artifactId>lancedb-core</artifactId>
-    <version>0.24.0</version>
+    <version>0.23.1</version>
 </dependency>
 ```

--- a/docs/src/styles/extra.css
+++ b/docs/src/styles/extra.css
@@ -85,26 +85,17 @@

 /* Header gradient (only header area) */
 .md-header {
-  background: linear-gradient(90deg, #e4d8f8 0%, #F0B7C1 45%, #E55A2B 100%);
+  background: linear-gradient(90deg, #3B2E58 0%, #F0B7C1 45%, #E55A2B 100%);
  box-shadow: inset 0 1px 0 rgba(255,255,255,0.08), 0 1px 0 rgba(0,0,0,0.08);
 }

-/* Improve brand title contrast on the lavender side */
-.md-header__title,
-.md-header__topic,
-.md-header__title .md-ellipsis,
-.md-header__topic .md-ellipsis {
-  color: #2b1b3a;
-  text-shadow: 0 1px 0 rgba(255, 255, 255, 0.25);
-}
-
 /* Same colors as header for tabs (that hold the text) */
 .md-tabs {
-  background: linear-gradient(90deg, #e4d8f8 0%, #F0B7C1 45%, #E55A2B 100%);
+  background: linear-gradient(90deg, #3B2E58 0%, #F0B7C1 45%, #E55A2B 100%);
 }

 /* Dark scheme variant */
 [data-md-color-scheme="slate"] .md-header,
 [data-md-color-scheme="slate"] .md-tabs {
-  background: linear-gradient(90deg, #e4d8f8 0%, #F0B7C1 45%, #E55A2B 100%);
+  background: linear-gradient(90deg, #3B2E58 0%, #F0B7C1 45%, #E55A2B 100%);
 }
--- a/java/lancedb-core/pom.xml
+++ b/java/lancedb-core/pom.xml
@@ -8,7 +8,7 @@
    <parent>
      <groupId>com.lancedb</groupId>
      <artifactId>lancedb-parent</artifactId>
-      <version>0.24.0-final.0</version>
+      <version>0.23.1-final.0</version>
      <relativePath>../pom.xml</relativePath>
    </parent>

--- a/java/pom.xml
+++ b/java/pom.xml
@@ -6,7 +6,7 @@

    <groupId>com.lancedb</groupId>
    <artifactId>lancedb-parent</artifactId>
-    <version>0.24.0-final.0</version>
+    <version>0.23.1-final.0</version>
    <packaging>pom</packaging>
    <name>${project.artifactId}</name>
    <description>LanceDB Java SDK Parent POM</description>
--- a/nodejs/Cargo.toml
+++ b/nodejs/Cargo.toml
@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.24.0"
+version = "0.23.1"
 license.workspace = true
 description.workspace = true
 repository.workspace = true
@@ -36,6 +36,6 @@ aws-lc-rs = "=1.13.0"
 napi-build = "2.1"

 [features]
-default = ["remote", "lancedb/aws", "lancedb/gcs", "lancedb/azure", "lancedb/dynamodb", "lancedb/oss", "lancedb/huggingface"]
+default = ["remote", "lancedb/default"]
 fp16kernels = ["lancedb/fp16kernels"]
 remote = ["lancedb/remote"]
--- a/nodejs/test/table.test.ts
+++ b/nodejs/test/table.test.ts
@@ -1520,9 +1520,9 @@ describe("when optimizing a dataset", () => {

  it("delete unverified", async () => {
    const version = await table.version();
-    const versionFile = `${tmpDir.name}/${table.name}.lance/_versions/${String(
-      18446744073709551615n - (BigInt(version) - 1n),
-    ).padStart(20, "0")}.manifest`;
+    const versionFile = `${tmpDir.name}/${table.name}.lance/_versions/${
+      version - 1
+    }.manifest`;
    fs.rmSync(versionFile);

    let stats = await table.optimize({ deleteUnverified: false });
--- a/nodejs/npm/darwin-arm64/package.json
+++ b/nodejs/npm/darwin-arm64/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",
--- a/nodejs/npm/darwin-x64/package.json
+++ b/nodejs/npm/darwin-x64/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-x64",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["darwin"],
 	"cpu": ["x64"],
 	"main": "lancedb.darwin-x64.node",
--- a/nodejs/npm/linux-arm64-gnu/package.json
+++ b/nodejs/npm/linux-arm64-gnu/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",
--- a/nodejs/npm/linux-arm64-musl/package.json
+++ b/nodejs/npm/linux-arm64-musl/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",
--- a/nodejs/npm/linux-x64-gnu/package.json
+++ b/nodejs/npm/linux-x64-gnu/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",
--- a/nodejs/npm/linux-x64-musl/package.json
+++ b/nodejs/npm/linux-x64-musl/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",
--- a/nodejs/npm/win32-arm64-msvc/package.json
+++ b/nodejs/npm/win32-arm64-msvc/package.json
@@ -1,6 +1,6 @@
 {
  "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.24.0",
+  "version": "0.23.1",
  "os": [
    "win32"
  ],
--- a/nodejs/npm/win32-x64-msvc/package.json
+++ b/nodejs/npm/win32-x64-msvc/package.json
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.24.0",
+	"version": "0.23.1",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",
--- a/nodejs/package-lock.json
+++ b/nodejs/package-lock.json
@@ -1,12 +1,12 @@
 {
  "name": "@lancedb/lancedb",
-  "version": "0.24.0",
+  "version": "0.23.1-beta.1",
  "lockfileVersion": 3,
  "requires": true,
  "packages": {
    "": {
      "name": "@lancedb/lancedb",
-      "version": "0.24.0",
+      "version": "0.23.1-beta.1",
      "cpu": [
        "x64",
        "arm64"
--- a/nodejs/package.json
+++ b/nodejs/package.json
@@ -11,7 +11,7 @@
    "ann"
  ],
  "private": false,
-  "version": "0.24.0",
+  "version": "0.23.1",
  "main": "dist/index.js",
  "exports": {
    ".": "./dist/index.js",
--- a/python/.bumpversion.toml
+++ b/python/.bumpversion.toml
@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.27.0"
+current_version = "0.26.1"
 parse = """(?x)
    (?P<major>0|[1-9]\\d*)\\.
    (?P<minor>0|[1-9]\\d*)\\.
--- a/python/Cargo.toml
+++ b/python/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.27.0"
+version = "0.26.1"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true
@@ -14,15 +14,15 @@ name = "_lancedb"
 crate-type = ["cdylib"]

 [dependencies]
-arrow = { version = "57.2", features = ["pyarrow"] }
+arrow = { version = "56.2", features = ["pyarrow"] }
 async-trait = "0.1"
 lancedb = { path = "../rust/lancedb", default-features = false }
 lance-core.workspace = true
 lance-namespace.workspace = true
 lance-io.workspace = true
 env_logger.workspace = true
-pyo3 = { version = "0.26", features = ["extension-module", "abi3-py39"] }
-pyo3-async-runtimes = { version = "0.26", features = [
+pyo3 = { version = "0.25", features = ["extension-module", "abi3-py39"] }
+pyo3-async-runtimes = { version = "0.25", features = [
    "attributes",
    "tokio-runtime",
 ] }
@@ -32,12 +32,12 @@ snafu.workspace = true
 tokio = { version = "1.40", features = ["sync"] }

 [build-dependencies]
-pyo3-build-config = { version = "0.26", features = [
+pyo3-build-config = { version = "0.25", features = [
    "extension-module",
    "abi3-py39",
 ] }

 [features]
-default = ["remote",  "lancedb/aws", "lancedb/gcs", "lancedb/azure", "lancedb/dynamodb", "lancedb/oss", "lancedb/huggingface"]
+default = ["remote", "lancedb/default"]
 fp16kernels = ["lancedb/fp16kernels"]
 remote = ["lancedb/remote"]
--- a/python/python/lancedb/init.py
+++ b/python/python/lancedb/init.py
@@ -13,7 +13,6 @@ __version__ = importlib.metadata.version("lancedb")

 from ._lancedb import connect as lancedb_connect
 from .common import URI, sanitize_uri
-from urllib.parse import urlparse
 from .db import AsyncConnection, DBConnection, LanceDBConnection
 from .io import StorageOptionsProvider
 from .remote import ClientConfig
@@ -29,39 +28,6 @@ from .namespace import (
 )


-def _check_s3_bucket_with_dots(
-    uri: str, storage_options: Optional[Dict[str, str]]
-) -> None:
-    """
-    Check if an S3 URI has a bucket name containing dots and warn if no region
-    is specified. S3 buckets with dots cannot use virtual-hosted-style URLs,
-    which breaks automatic region detection.
-
-    See: https://github.com/lancedb/lancedb/issues/1898
-    """
-    if not isinstance(uri, str) or not uri.startswith("s3://"):
-        return
-
-    parsed = urlparse(uri)
-    bucket = parsed.netloc
-
-    if "." not in bucket:
-        return
-
-    # Check if region is provided in storage_options
-    region_keys = {"region", "aws_region"}
-    has_region = storage_options and any(k in storage_options for k in region_keys)
-
-    if not has_region:
-        raise ValueError(
-            f"S3 bucket name '{bucket}' contains dots, which prevents automatic "
-            f"region detection. Please specify the region explicitly via "
-            f"storage_options={{'region': '<your-region>'}} or "
-            f"storage_options={{'aws_region': '<your-region>'}}. "
-            f"See https://github.com/lancedb/lancedb/issues/1898 for details."
-        )
-
-
 def connect(
    uri: URI,
    *,
@@ -155,11 +121,9 @@ def connect(
            storage_options=storage_options,
            **kwargs,
        )
-    _check_s3_bucket_with_dots(str(uri), storage_options)

    if kwargs:
        raise ValueError(f"Unknown keyword arguments: {kwargs}")
-
    return LanceDBConnection(
        uri,
        read_consistency_interval=read_consistency_interval,
@@ -247,8 +211,6 @@ async def connect_async(
    if isinstance(client_config, dict):
        client_config = ClientConfig(**client_config)

-    _check_s3_bucket_with_dots(str(uri), storage_options)
-
    return AsyncConnection(
        await lancedb_connect(
            sanitize_uri(uri),
--- a/python/python/lancedb/_lancedb.pyi
+++ b/python/python/lancedb/_lancedb.pyi
@@ -179,7 +179,6 @@ class Table:
        cleanup_since_ms: Optional[int] = None,
        delete_unverified: Optional[bool] = None,
    ) -> OptimizeStats: ...
-    async def uri(self) -> str: ...
    @property
    def tags(self) -> Tags: ...
    def query(self) -> Query: ...
--- a/python/python/lancedb/embeddings/colpali.py
+++ b/python/python/lancedb/embeddings/colpali.py
@@ -9,8 +9,6 @@ import numpy as np
 import io
 import warnings

-from pydantic import Field
-
 from ..util import attempt_import_or_raise
 from .base import EmbeddingFunction
 from .registry import register
@@ -28,7 +26,7 @@ class ColPaliEmbeddings(EmbeddingFunction):

    Parameters
    ----------
-    colpali_model_name : str
+    model_name : str
        The name of the model to use (e.g., "Metric-AI/ColQwen2.5-3b-multilingual-v1.0")
        Supports models based on these engines:
        - ColPali: "vidore/colpali-v1.3" and others
@@ -59,10 +57,7 @@ class ColPaliEmbeddings(EmbeddingFunction):
        useful for large models that do not fit in memory.
    """

-    colpali_model_name: str = Field(
-        default="Metric-AI/ColQwen2.5-3b-multilingual-v1.0",
-        validation_alias="model_name",
-    )
+    model_name: str = "Metric-AI/ColQwen2.5-3b-multilingual-v1.0"
    device: str = "auto"
    dtype: str = "bfloat16"
    use_token_pooling: bool = True
@@ -112,7 +107,7 @@ class ColPaliEmbeddings(EmbeddingFunction):
            self._processor,
            self._token_pooler,
        ) = self._load_model(
-            self.colpali_model_name,
+            self.model_name,
            dtype,
            device,
            self.pooling_strategy,
--- a/python/python/lancedb/embeddings/siglip.py
+++ b/python/python/lancedb/embeddings/siglip.py
@@ -10,7 +10,7 @@ import urllib.parse as urlparse
 import numpy as np
 import pyarrow as pa
 from tqdm import tqdm
-from pydantic import Field, PrivateAttr
+from pydantic import PrivateAttr

 from ..util import attempt_import_or_raise
 from .base import EmbeddingFunction
@@ -24,10 +24,7 @@ if TYPE_CHECKING:

@register("siglip")
 class SigLipEmbeddings(EmbeddingFunction):
-    siglip_model_name: str = Field(
-        default="google/siglip-base-patch16-224",
-        validation_alias="model_name",
-    )
+    model_name: str = "google/siglip-base-patch16-224"
    device: str = "cpu"
    batch_size: int = 64
    normalize: bool = True
@@ -42,10 +39,8 @@ class SigLipEmbeddings(EmbeddingFunction):
        transformers = attempt_import_or_raise("transformers")
        self._torch = attempt_import_or_raise("torch")

-        self._processor = transformers.AutoProcessor.from_pretrained(
-            self.siglip_model_name
-        )
-        self._model = transformers.SiglipModel.from_pretrained(self.siglip_model_name)
+        self._processor = transformers.AutoProcessor.from_pretrained(self.model_name)
+        self._model = transformers.SiglipModel.from_pretrained(self.model_name)
        self._model.to(self.device)
        self._model.eval()
        self._ndims = None
--- a/python/python/lancedb/embeddings/voyageai.py
+++ b/python/python/lancedb/embeddings/voyageai.py
@@ -2,7 +2,7 @@
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
 import base64
 import os
-from typing import ClassVar, TYPE_CHECKING, List, Union, Any, Generator, Optional
+from typing import ClassVar, TYPE_CHECKING, List, Union, Any, Generator

 from pathlib import Path
 from urllib.parse import urlparse
@@ -45,29 +45,11 @@ def is_valid_url(text):
        return False


-VIDEO_EXTENSIONS = {".mp4", ".webm", ".mov", ".avi", ".mkv", ".m4v", ".gif"}
-
-
-def is_video_url(url: str) -> bool:
-    """Check if URL points to a video file based on extension."""
-    parsed = urlparse(url)
-    path = parsed.path.lower()
-    return any(path.endswith(ext) for ext in VIDEO_EXTENSIONS)
-
-
-def is_video_path(path: Path) -> bool:
-    """Check if file path is a video file based on extension."""
-    return path.suffix.lower() in VIDEO_EXTENSIONS
-
-
 def transform_input(input_data: Union[str, bytes, Path]):
    PIL = attempt_import_or_raise("PIL", "pillow")
    if isinstance(input_data, str):
        if is_valid_url(input_data):
-            if is_video_url(input_data):
-                content = {"type": "video_url", "video_url": input_data}
-            else:
-                content = {"type": "image_url", "image_url": input_data}
+            content = {"type": "image_url", "image_url": input_data}
        else:
            content = {"type": "text", "text": input_data}
    elif isinstance(input_data, PIL.Image.Image):
@@ -88,24 +70,14 @@ def transform_input(input_data: Union[str, bytes, Path]):
            "image_base64": "data:image/jpeg;base64," + img_str,
        }
    elif isinstance(input_data, Path):
-        if is_video_path(input_data):
-            # Read video file and encode as base64
-            with open(input_data, "rb") as f:
-                video_bytes = f.read()
-            video_str = base64.b64encode(video_bytes).decode("utf-8")
-            content = {
-                "type": "video_base64",
-                "video_base64": video_str,
-            }
-        else:
-            img = PIL.Image.open(input_data)
-            buffered = BytesIO()
-            img.save(buffered, format="JPEG")
-            img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
-            content = {
-                "type": "image_base64",
-                "image_base64": "data:image/jpeg;base64," + img_str,
-            }
+        img = PIL.Image.open(input_data)
+        buffered = BytesIO()
+        img.save(buffered, format="JPEG")
+        img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")
+        content = {
+            "type": "image_base64",
+            "image_base64": "data:image/jpeg;base64," + img_str,
+        }
    else:
        raise ValueError("Each input should be either str, bytes, Path or Image.")

@@ -119,8 +91,6 @@ def sanitize_multimodal_input(inputs: Union[TEXT, IMAGES]) -> List[Any]:
    PIL = attempt_import_or_raise("PIL", "pillow")
    if isinstance(inputs, (str, bytes, Path, PIL.Image.Image)):
        inputs = [inputs]
-    elif isinstance(inputs, list):
-        pass  # Already a list, use as-is
    elif isinstance(inputs, pa.Array):
        inputs = inputs.to_pylist()
    elif isinstance(inputs, pa.ChunkedArray):
@@ -173,16 +143,11 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
            * voyage-3
            * voyage-3-lite
            * voyage-multimodal-3
-            * voyage-multimodal-3.5
            * voyage-finance-2
            * voyage-multilingual-2
            * voyage-law-2
            * voyage-code-2

-    output_dimension: int, optional
-        The output dimension for models that support flexible dimensions.
-        Currently only voyage-multimodal-3.5 supports this feature.
-        Valid options: 256, 512, 1024 (default), 2048.

    Examples
    --------
@@ -210,10 +175,7 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
    """

    name: str
-    output_dimension: Optional[int] = None
    client: ClassVar = None
-    _FLEXIBLE_DIM_MODELS: ClassVar[list] = ["voyage-multimodal-3.5"]
-    _VALID_DIMENSIONS: ClassVar[list] = [256, 512, 1024, 2048]
    text_embedding_models: list = [
        "voyage-3.5",
        "voyage-3.5-lite",
@@ -224,7 +186,7 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
        "voyage-law-2",
        "voyage-code-2",
    ]
-    multimodal_embedding_models: list = ["voyage-multimodal-3", "voyage-multimodal-3.5"]
+    multimodal_embedding_models: list = ["voyage-multimodal-3"]
    contextual_embedding_models: list = ["voyage-context-3"]

    def _is_multimodal_model(self, model_name: str):
@@ -236,17 +198,6 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
        return model_name in self.contextual_embedding_models or "context" in model_name

    def ndims(self):
-        # Handle flexible dimension models
-        if self.name in self._FLEXIBLE_DIM_MODELS:
-            if self.output_dimension is not None:
-                if self.output_dimension not in self._VALID_DIMENSIONS:
-                    raise ValueError(
-                        f"Invalid output_dimension {self.output_dimension} "
-                        f"for {self.name}. Valid options: {self._VALID_DIMENSIONS}"
-                    )
-                return self.output_dimension
-            return 1024  # default dimension
-
        if self.name == "voyage-3-lite":
            return 512
        elif self.name == "voyage-code-2":
@@ -260,17 +211,12 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
            "voyage-finance-2",
            "voyage-multilingual-2",
            "voyage-law-2",
+            "voyage-multimodal-3",
        ]:
            return 1024
        else:
            raise ValueError(f"Model {self.name} not supported")

-    def _get_multimodal_kwargs(self, **kwargs):
-        """Get kwargs for multimodal embed call, including output_dimension if set."""
-        if self.name in self._FLEXIBLE_DIM_MODELS and self.output_dimension is not None:
-            kwargs["output_dimension"] = self.output_dimension
-        return kwargs
-
    def compute_query_embeddings(
        self, query: Union[str, "PIL.Image.Image"], *args, **kwargs
    ) -> List[np.ndarray]:
@@ -288,7 +234,6 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
        """
        client = VoyageAIEmbeddingFunction._get_client()
        if self._is_multimodal_model(self.name):
-            kwargs = self._get_multimodal_kwargs(**kwargs)
            result = client.multimodal_embed(
                inputs=[[query]], model=self.name, input_type="query", **kwargs
            )
@@ -330,7 +275,6 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
            )
            if has_images:
                # Use non-batched API for images
-                kwargs = self._get_multimodal_kwargs(**kwargs)
                result = client.multimodal_embed(
                    inputs=sanitized, model=self.name, input_type="document", **kwargs
                )
@@ -413,7 +357,6 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
            callable: A function that takes a batch of texts and returns embeddings.
        """
        if self._is_multimodal_model(self.name):
-            multimodal_kwargs = self._get_multimodal_kwargs(**kwargs)

            def embed_batch(batch: List[str]) -> List[np.array]:
                batch_inputs = sanitize_multimodal_input(batch)
@@ -421,7 +364,7 @@ class VoyageAIEmbeddingFunction(EmbeddingFunction):
                    inputs=batch_inputs,
                    model=self.name,
                    input_type=input_type,
-                    **multimodal_kwargs,
+                    **kwargs,
                )
                return result.embeddings

--- a/python/python/lancedb/query.py
+++ b/python/python/lancedb/query.py
@@ -961,27 +961,22 @@ class LanceQueryBuilder(ABC):
        >>> query = [100, 100]
        >>> plan = table.search(query).analyze_plan()
        >>> print(plan)  # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
-        AnalyzeExec verbose=true, elapsed=..., metrics=...
-          TracedExec, elapsed=..., metrics=...
-            ProjectionExec: elapsed=..., expr=[...],
-            metrics=[output_rows=..., elapsed_compute=..., output_bytes=...]
-              GlobalLimitExec: elapsed=..., skip=0, fetch=10,
-              metrics=[output_rows=..., elapsed_compute=..., output_bytes=...]
-                FilterExec: elapsed=..., _distance@2 IS NOT NULL, metrics=[...]
-                  SortExec: elapsed=..., TopK(fetch=10), expr=[...],
+        AnalyzeExec verbose=true, metrics=[], cumulative_cpu=...
+          TracedExec, metrics=[], cumulative_cpu=...
+            ProjectionExec: expr=[...], metrics=[...], cumulative_cpu=...
+              GlobalLimitExec: skip=0, fetch=10, metrics=[...], cumulative_cpu=...
+                FilterExec: _distance@2 IS NOT NULL,
+                metrics=[output_rows=..., elapsed_compute=...], cumulative_cpu=...
+                  SortExec: TopK(fetch=10), expr=[...],
                  preserve_partitioning=[...],
-                  metrics=[output_rows=..., elapsed_compute=...,
-                  output_bytes=..., row_replacements=...]
-                    KNNVectorDistance: elapsed=..., metric=l2,
-                    metrics=[output_rows=..., elapsed_compute=...,
-                    output_bytes=..., output_batches=...]
-                      LanceRead: elapsed=..., uri=..., projection=[vector],
-                      num_fragments=..., range_before=None, range_after=None,
-                      row_id=true, row_addr=false,
-                      full_filter=--, refine_filter=--,
-                      metrics=[output_rows=..., elapsed_compute=..., output_bytes=...,
-                      fragments_scanned=..., ranges_scanned=1, rows_scanned=1,
-                      bytes_read=..., iops=..., requests=..., task_wait_time=...]
+                  metrics=[output_rows=..., elapsed_compute=..., row_replacements=...],
+                  cumulative_cpu=...
+                    KNNVectorDistance: metric=l2,
+                    metrics=[output_rows=..., elapsed_compute=..., output_batches=...],
+                    cumulative_cpu=...
+                      LanceRead: uri=..., projection=[vector], ...
+                      metrics=[output_rows=..., elapsed_compute=...,
+                      bytes_read=..., iops=..., requests=...], cumulative_cpu=...

        Returns
        -------
--- a/python/python/lancedb/remote/db.py
+++ b/python/python/lancedb/remote/db.py
@@ -384,7 +384,6 @@ class RemoteDBConnection(DBConnection):
        on_bad_vectors: str = "error",
        fill_value: float = 0.0,
        mode: Optional[str] = None,
-        exist_ok: bool = False,
        embedding_functions: Optional[List[EmbeddingFunctionConfig]] = None,
        *,
        namespace: Optional[List[str]] = None,
@@ -413,12 +412,6 @@ class RemoteDBConnection(DBConnection):
            - pyarrow.Schema

            - [LanceModel][lancedb.pydantic.LanceModel]
-        mode: str, default "create"
-            The mode to use when creating the table.
-            Can be either "create", "overwrite", or "exist_ok".
-        exist_ok: bool, default False
-            If exist_ok is True, and mode is None or "create", mode will be changed
-            to "exist_ok".
        on_bad_vectors: str, default "error"
            What to do if any of the vectors are not the same size or contains NaNs.
            One of "error", "drop", "fill".
@@ -490,11 +483,6 @@ class RemoteDBConnection(DBConnection):
        LanceTable(table4)

        """
-        if exist_ok:
-            if mode == "create":
-                mode = "exist_ok"
-            elif not mode:
-                mode = "exist_ok"
        if namespace is None:
            namespace = []
        validate_table_name(name)
--- a/python/python/lancedb/remote/table.py
+++ b/python/python/lancedb/remote/table.py
@@ -18,17 +18,7 @@ from lancedb._lancedb import (
    UpdateResult,
 )
 from lancedb.embeddings.base import EmbeddingFunctionConfig
-from lancedb.index import (
-    FTS,
-    BTree,
-    Bitmap,
-    HnswSq,
-    IvfFlat,
-    IvfPq,
-    IvfRq,
-    IvfSq,
-    LabelList,
-)
+from lancedb.index import FTS, BTree, Bitmap, HnswSq, IvfFlat, IvfPq, IvfSq, LabelList
 from lancedb.remote.db import LOOP
 import pyarrow as pa

@@ -275,12 +265,6 @@ class RemoteTable(Table):
                num_sub_vectors=num_sub_vectors,
                num_bits=num_bits,
            )
-        elif index_type == "IVF_RQ":
-            config = IvfRq(
-                distance_type=metric,
-                num_partitions=num_partitions,
-                num_bits=num_bits,
-            )
        elif index_type == "IVF_SQ":
            config = IvfSq(distance_type=metric, num_partitions=num_partitions)
        elif index_type == "IVF_HNSW_PQ":
@@ -295,8 +279,7 @@ class RemoteTable(Table):
        else:
            raise ValueError(
                f"Unknown vector index type: {index_type}. Valid options are"
-                " 'IVF_FLAT', 'IVF_PQ', 'IVF_RQ', 'IVF_SQ',"
-                " 'IVF_HNSW_PQ', 'IVF_HNSW_SQ'"
+                " 'IVF_FLAT', 'IVF_SQ', 'IVF_PQ', 'IVF_HNSW_PQ', 'IVF_HNSW_SQ'"
            )

        LOOP.run(
@@ -655,14 +638,6 @@ class RemoteTable(Table):
    def stats(self):
        return LOOP.run(self._table.stats())

-    @property
-    def uri(self) -> str:
-        """The table URI (storage location).
-
-        For remote tables, this fetches the location from the server via describe.
-        """
-        return LOOP.run(self._table.uri())
-
    def take_offsets(self, offsets: list[int]) -> LanceTakeQueryBuilder:
        return LanceTakeQueryBuilder(self._table.take_offsets(offsets))

--- a/python/python/lancedb/table.py
+++ b/python/python/lancedb/table.py
@@ -2218,10 +2218,6 @@ class LanceTable(Table):
    def stats(self) -> TableStatistics:
        return LOOP.run(self._table.stats())

-    @property
-    def uri(self) -> str:
-        return LOOP.run(self._table.uri())
-
    def create_scalar_index(
        self,
        column: str,
@@ -3610,20 +3606,6 @@ class AsyncTable:
        """
        return await self._inner.stats()

-    async def uri(self) -> str:
-        """
-        Get the table URI (storage location).
-
-        For remote tables, this fetches the location from the server via describe.
-        For local tables, this returns the dataset URI.
-
-        Returns
-        -------
-        str
-            The full storage location of the table (e.g., S3/GCS path).
-        """
-        return await self._inner.uri()
-
    async def add(
        self,
        data: DATA,
--- a/python/python/tests/conftest.py
+++ b/python/python/tests/conftest.py
@@ -2,27 +2,12 @@
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors

 from datetime import timedelta
-
 from lancedb.db import AsyncConnection, DBConnection
 import lancedb
 import pytest
 import pytest_asyncio


-def pandas_string_type():
-    """Return the PyArrow string type that pandas uses for string columns.
-
-    pandas 3.0+ uses large_string for string columns, pandas 2.x uses string.
-    """
-    import pandas as pd
-    import pyarrow as pa
-
-    version = tuple(int(x) for x in pd.__version__.split(".")[:2])
-    if version >= (3, 0):
-        return pa.large_utf8()
-    return pa.utf8()
-
-
 # Use an in-memory database for most tests.
@pytest.fixture
 def mem_db() -> DBConnection:
--- a/python/python/tests/test_db.py
+++ b/python/python/tests/test_db.py
@@ -268,8 +268,6 @@ async def test_create_table_from_iterator_async(mem_db_async: lancedb.AsyncConne


 def test_create_exist_ok(tmp_db: lancedb.DBConnection):
-    from conftest import pandas_string_type
-
    data = pd.DataFrame(
        {
            "vector": [[3.1, 4.1], [5.9, 26.5]],
@@ -288,11 +286,10 @@ def test_create_exist_ok(tmp_db: lancedb.DBConnection):
    assert tbl.schema == tbl2.schema
    assert len(tbl) == len(tbl2)

-    # pandas 3.0+ uses large_string, pandas 2.x uses string
    schema = pa.schema(
        [
            pa.field("vector", pa.list_(pa.float32(), list_size=2)),
-            pa.field("item", pandas_string_type()),
+            pa.field("item", pa.utf8()),
            pa.field("price", pa.float64()),
        ]
    )
@@ -302,7 +299,7 @@ def test_create_exist_ok(tmp_db: lancedb.DBConnection):
    bad_schema = pa.schema(
        [
            pa.field("vector", pa.list_(pa.float32(), list_size=2)),
-            pa.field("item", pandas_string_type()),
+            pa.field("item", pa.utf8()),
            pa.field("price", pa.float64()),
            pa.field("extra", pa.float32()),
        ]
@@ -368,8 +365,6 @@ async def test_create_mode_async(tmp_db_async: lancedb.AsyncConnection):

@pytest.mark.asyncio
 async def test_create_exist_ok_async(tmp_db_async: lancedb.AsyncConnection):
-    from conftest import pandas_string_type
-
    data = pd.DataFrame(
        {
            "vector": [[3.1, 4.1], [5.9, 26.5]],
@@ -387,11 +382,10 @@ async def test_create_exist_ok_async(tmp_db_async: lancedb.AsyncConnection):
    assert tbl.name == tbl2.name
    assert await tbl.schema() == await tbl2.schema()

-    # pandas 3.0+ uses large_string, pandas 2.x uses string
    schema = pa.schema(
        [
            pa.field("vector", pa.list_(pa.float32(), list_size=2)),
-            pa.field("item", pandas_string_type()),
+            pa.field("item", pa.utf8()),
            pa.field("price", pa.float64()),
        ]
    )
@@ -601,8 +595,6 @@ def test_open_table_sync(tmp_db: lancedb.DBConnection):

@pytest.mark.asyncio
 async def test_open_table(tmp_path):
-    from conftest import pandas_string_type
-
    db = await lancedb.connect_async(tmp_path)
    data = pd.DataFrame(
        {
@@ -622,11 +614,10 @@ async def test_open_table(tmp_path):
        )
        is not None
    )
-    # pandas 3.0+ uses large_string, pandas 2.x uses string
    assert await tbl.schema() == pa.schema(
        {
            "vector": pa.list_(pa.float32(), list_size=2),
-            "item": pandas_string_type(),
+            "item": pa.utf8(),
            "price": pa.float64(),
        }
    )
--- a/python/python/tests/test_embeddings_slow.py
+++ b/python/python/tests/test_embeddings_slow.py
@@ -613,133 +613,6 @@ def test_voyageai_multimodal_embedding_text_function():
    assert len(tbl.to_pandas()["vector"][0]) == voyageai.ndims()


-@pytest.mark.slow
-@pytest.mark.skipif(
-    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
-)
-def test_voyageai_multimodal_35_embedding_function():
-    """Test voyage-multimodal-3.5 model with text input."""
-    voyageai = (
-        get_registry()
-        .get("voyageai")
-        .create(name="voyage-multimodal-3.5", max_retries=0)
-    )
-
-    class TextModel(LanceModel):
-        text: str = voyageai.SourceField()
-        vector: Vector(voyageai.ndims()) = voyageai.VectorField()
-
-    df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
-    db = lancedb.connect("~/lancedb")
-    tbl = db.create_table("test_multimodal_35", schema=TextModel, mode="overwrite")
-
-    tbl.add(df)
-    assert len(tbl.to_pandas()["vector"][0]) == voyageai.ndims()
-    assert voyageai.ndims() == 1024
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(
-    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
-)
-def test_voyageai_multimodal_35_flexible_dimensions():
-    """Test voyage-multimodal-3.5 model with custom output dimension."""
-    voyageai = (
-        get_registry()
-        .get("voyageai")
-        .create(name="voyage-multimodal-3.5", output_dimension=512, max_retries=0)
-    )
-
-    class TextModel(LanceModel):
-        text: str = voyageai.SourceField()
-        vector: Vector(voyageai.ndims()) = voyageai.VectorField()
-
-    assert voyageai.ndims() == 512
-
-    df = pd.DataFrame({"text": ["hello world", "goodbye world"]})
-    db = lancedb.connect("~/lancedb")
-    tbl = db.create_table("test_multimodal_35_dim", schema=TextModel, mode="overwrite")
-
-    tbl.add(df)
-    assert len(tbl.to_pandas()["vector"][0]) == 512
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(
-    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
-)
-def test_voyageai_multimodal_35_image_embedding():
-    """Test voyage-multimodal-3.5 model with image input."""
-    voyageai = (
-        get_registry()
-        .get("voyageai")
-        .create(name="voyage-multimodal-3.5", max_retries=0)
-    )
-
-    class Images(LanceModel):
-        label: str
-        image_uri: str = voyageai.SourceField()
-        vector: Vector(voyageai.ndims()) = voyageai.VectorField()
-
-    db = lancedb.connect("~/lancedb")
-    table = db.create_table(
-        "test_multimodal_35_images", schema=Images, mode="overwrite"
-    )
-    labels = ["cat", "dog"]
-    uris = [
-        "http://farm1.staticflickr.com/53/167798175_7c7845bbbd_z.jpg",
-        "http://farm9.staticflickr.com/8387/8602747737_2e5c2a45d4_z.jpg",
-    ]
-    table.add(pd.DataFrame({"label": labels, "image_uri": uris}))
-    assert len(table.to_pandas()["vector"][0]) == voyageai.ndims()
-    assert voyageai.ndims() == 1024
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(
-    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
-)
-@pytest.mark.parametrize("dimension", [256, 512, 1024, 2048])
-def test_voyageai_multimodal_35_all_dimensions(dimension):
-    """Test voyage-multimodal-3.5 model with all valid output dimensions."""
-    voyageai = (
-        get_registry()
-        .get("voyageai")
-        .create(name="voyage-multimodal-3.5", output_dimension=dimension, max_retries=0)
-    )
-
-    assert voyageai.ndims() == dimension
-
-    class TextModel(LanceModel):
-        text: str = voyageai.SourceField()
-        vector: Vector(voyageai.ndims()) = voyageai.VectorField()
-
-    df = pd.DataFrame({"text": ["hello world"]})
-    db = lancedb.connect("~/lancedb")
-    tbl = db.create_table(
-        f"test_multimodal_35_dim_{dimension}", schema=TextModel, mode="overwrite"
-    )
-
-    tbl.add(df)
-    assert len(tbl.to_pandas()["vector"][0]) == dimension
-
-
-@pytest.mark.slow
-@pytest.mark.skipif(
-    os.environ.get("VOYAGE_API_KEY") is None, reason="VOYAGE_API_KEY not set"
-)
-def test_voyageai_multimodal_35_invalid_dimension():
-    """Test voyage-multimodal-3.5 model raises error for invalid output dimension."""
-    with pytest.raises(ValueError, match="Invalid output_dimension"):
-        voyageai = (
-            get_registry()
-            .get("voyageai")
-            .create(name="voyage-multimodal-3.5", output_dimension=999, max_retries=0)
-        )
-        # ndims() is where the validation happens
-        voyageai.ndims()
-
-
@pytest.mark.slow
@pytest.mark.skipif(
    importlib.util.find_spec("colpali_engine") is None,
--- a/python/python/tests/test_namespace_integration.py
+++ b/python/python/tests/test_namespace_integration.py
@@ -26,8 +26,6 @@ import pytest
 from lance_namespace import (
    CreateEmptyTableRequest,
    CreateEmptyTableResponse,
-    DeclareTableRequest,
-    DeclareTableResponse,
    DescribeTableRequest,
    DescribeTableResponse,
    LanceNamespace,
@@ -162,19 +160,6 @@ class TrackingNamespace(LanceNamespace):

        return modified

-    def declare_table(self, request: DeclareTableRequest) -> DeclareTableResponse:
-        """Track declare_table calls and inject rotating credentials."""
-        with self.lock:
-            self.create_call_count += 1
-            count = self.create_call_count
-
-        response = self.inner.declare_table(request)
-        response.storage_options = self._modify_storage_options(
-            response.storage_options, count
-        )
-
-        return response
-
    def create_empty_table(
        self, request: CreateEmptyTableRequest
    ) -> CreateEmptyTableResponse:
--- a/python/python/tests/test_remote_db.py
+++ b/python/python/tests/test_remote_db.py
@@ -168,42 +168,6 @@ def test_table_len_sync():
        assert len(table) == 1


-def test_create_table_exist_ok():
-    def handler(request):
-        if request.path == "/v1/table/test/create/?mode=exist_ok":
-            request.send_response(200)
-            request.send_header("Content-Type", "application/json")
-            request.end_headers()
-            request.wfile.write(b"{}")
-        else:
-            request.send_response(404)
-            request.end_headers()
-
-    with mock_lancedb_connection(handler) as db:
-        table = db.create_table("test", [{"id": 1}], exist_ok=True)
-        assert table is not None
-
-    with mock_lancedb_connection(handler) as db:
-        table = db.create_table("test", [{"id": 1}], mode="create", exist_ok=True)
-        assert table is not None
-
-
-def test_create_table_exist_ok_with_mode_overwrite():
-    def handler(request):
-        if request.path == "/v1/table/test/create/?mode=overwrite":
-            request.send_response(200)
-            request.send_header("Content-Type", "application/json")
-            request.end_headers()
-            request.wfile.write(b"{}")
-        else:
-            request.send_response(404)
-            request.end_headers()
-
-    with mock_lancedb_connection(handler) as db:
-        table = db.create_table("test", [{"id": 1}], mode="overwrite", exist_ok=True)
-        assert table is not None
-
-
@pytest.mark.asyncio
 async def test_http_error():
    request_id_holder = {"request_id": None}
@@ -601,6 +565,7 @@ def test_head():
 def test_query_sync_minimal():
    def handler(body):
        assert body == {
+            "distance_type": "l2",
            "k": 10,
            "prefilter": True,
            "refine_factor": None,
@@ -684,6 +649,7 @@ def test_query_sync_maximal():
 def test_query_sync_nprobes():
    def handler(body):
        assert body == {
+            "distance_type": "l2",
            "k": 10,
            "prefilter": True,
            "fast_search": True,
@@ -713,6 +679,7 @@ def test_query_sync_nprobes():
 def test_query_sync_no_max_nprobes():
    def handler(body):
        assert body == {
+            "distance_type": "l2",
            "k": 10,
            "prefilter": True,
            "fast_search": True,
@@ -835,6 +802,7 @@ def test_query_sync_hybrid():
        else:
            # Vector query
            assert body == {
+                "distance_type": "l2",
                "k": 42,
                "prefilter": True,
                "refine_factor": None,
--- a/python/python/tests/test_s3_bucket_dots.py
+++ b/python/python/tests/test_s3_bucket_dots.py
@@ -1,68 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-"""
-Tests for S3 bucket names containing dots.
-
-Related issue: https://github.com/lancedb/lancedb/issues/1898
-
-These tests validate the early error checking for S3 bucket names with dots.
-No actual S3 connection is made - validation happens before connection.
-"""
-
-import pytest
-import lancedb
-
-# Test URIs
-BUCKET_WITH_DOTS = "s3://my.bucket.name/path"
-BUCKET_WITH_DOTS_AND_REGION = ("s3://my.bucket.name", {"region": "us-east-1"})
-BUCKET_WITH_DOTS_AND_AWS_REGION = ("s3://my.bucket.name", {"aws_region": "us-east-1"})
-BUCKET_WITHOUT_DOTS = "s3://my-bucket/path"
-
-
-class TestS3BucketWithDotsSync:
-    """Tests for connect()."""
-
-    def test_bucket_with_dots_requires_region(self):
-        with pytest.raises(ValueError, match="contains dots"):
-            lancedb.connect(BUCKET_WITH_DOTS)
-
-    def test_bucket_with_dots_and_region_passes(self):
-        uri, opts = BUCKET_WITH_DOTS_AND_REGION
-        db = lancedb.connect(uri, storage_options=opts)
-        assert db is not None
-
-    def test_bucket_with_dots_and_aws_region_passes(self):
-        uri, opts = BUCKET_WITH_DOTS_AND_AWS_REGION
-        db = lancedb.connect(uri, storage_options=opts)
-        assert db is not None
-
-    def test_bucket_without_dots_passes(self):
-        db = lancedb.connect(BUCKET_WITHOUT_DOTS)
-        assert db is not None
-
-
-class TestS3BucketWithDotsAsync:
-    """Tests for connect_async()."""
-
-    @pytest.mark.asyncio
-    async def test_bucket_with_dots_requires_region(self):
-        with pytest.raises(ValueError, match="contains dots"):
-            await lancedb.connect_async(BUCKET_WITH_DOTS)
-
-    @pytest.mark.asyncio
-    async def test_bucket_with_dots_and_region_passes(self):
-        uri, opts = BUCKET_WITH_DOTS_AND_REGION
-        db = await lancedb.connect_async(uri, storage_options=opts)
-        assert db is not None
-
-    @pytest.mark.asyncio
-    async def test_bucket_with_dots_and_aws_region_passes(self):
-        uri, opts = BUCKET_WITH_DOTS_AND_AWS_REGION
-        db = await lancedb.connect_async(uri, storage_options=opts)
-        assert db is not None
-
-    @pytest.mark.asyncio
-    async def test_bucket_without_dots_passes(self):
-        db = await lancedb.connect_async(BUCKET_WITHOUT_DOTS)
-        assert db is not None
--- a/python/python/tests/test_table.py
+++ b/python/python/tests/test_table.py
@@ -1880,13 +1880,8 @@ async def test_optimize_delete_unverified(tmp_db_async: AsyncConnection, tmp_pat
        ],
    )
    version = await table.version()
-    assert version == 2
-
-    # By removing a manifest file, we make the data files we just inserted unverified
-    version_name = 18446744073709551615 - (version - 1)
-    path = tmp_path / "test.lance" / "_versions" / f"{version_name:020}.manifest"
+    path = tmp_path / "test.lance" / "_versions" / f"{version - 1}.manifest"
    os.remove(path)
-
    stats = await table.optimize(delete_unverified=False)
    assert stats.prune.old_versions_removed == 0
    stats = await table.optimize(
@@ -1972,9 +1967,3 @@ def test_add_table_with_empty_embeddings(tmp_path):
        on_bad_vectors="drop",
    )
    assert table.count_rows() == 1
-
-
-def test_table_uri(tmp_path):
-    db = lancedb.connect(tmp_path)
-    table = db.create_table("my_table", data=[{"x": 0}])
-    assert table.uri == str(tmp_path / "my_table.lance")
--- a/python/python/tests/test_util.py
+++ b/python/python/tests/test_util.py
@@ -528,19 +528,12 @@ def test_sanitize_data(
        else:
            expected_schema = schema
    else:
-        from conftest import pandas_string_type
-
-        # polars uses large_string, pandas 3.0+ uses large_string, others use string
-        if isinstance(data, pl.DataFrame):
-            text_type = pa.large_utf8()
-        elif isinstance(data, pd.DataFrame):
-            text_type = pandas_string_type()
-        else:
-            text_type = pa.string()
        expected_schema = pa.schema(
            {
                "id": pa.int64(),
-                "text": text_type,
+                "text": pa.large_utf8()
+                if isinstance(data, pl.DataFrame)
+                else pa.string(),
                "vector": pa.list_(pa.float32(), 10),
            }
        )
--- a/python/src/arrow.rs
+++ b/python/src/arrow.rs
@@ -10,7 +10,8 @@ use arrow::{
 use futures::stream::StreamExt;
 use lancedb::arrow::SendableRecordBatchStream;
 use pyo3::{
-    exceptions::PyStopAsyncIteration, pyclass, pymethods, Bound, Py, PyAny, PyRef, PyResult, Python,
+    exceptions::PyStopAsyncIteration, pyclass, pymethods, Bound, PyAny, PyObject, PyRef, PyResult,
+    Python,
 };
 use pyo3_async_runtimes::tokio::future_into_py;

@@ -35,11 +36,8 @@ impl RecordBatchStream {
 #[pymethods]
 impl RecordBatchStream {
    #[getter]
-    pub fn schema(&self, py: Python) -> PyResult<Py<PyAny>> {
-        (*self.schema)
-            .clone()
-            .into_pyarrow(py)
-            .map(|obj| obj.unbind())
+    pub fn schema(&self, py: Python) -> PyResult<PyObject> {
+        (*self.schema).clone().into_pyarrow(py)
    }

    pub fn __aiter__(self_: PyRef<'_, Self>) -> PyRef<'_, Self> {
@@ -55,12 +53,7 @@ impl RecordBatchStream {
                .next()
                .await
                .ok_or_else(|| PyStopAsyncIteration::new_err(""))?;
-            Python::attach(|py| {
-                inner_next
-                    .infer_error()?
-                    .to_pyarrow(py)
-                    .map(|obj| obj.unbind())
-            })
+            Python::with_gil(|py| inner_next.infer_error()?.to_pyarrow(py))
        })
    }
 }
--- a/python/src/connection.rs
+++ b/python/src/connection.rs
@@ -12,7 +12,7 @@ use pyo3::{
    exceptions::{PyRuntimeError, PyValueError},
    pyclass, pyfunction, pymethods,
    types::{PyDict, PyDictMethods},
-    Bound, FromPyObject, Py, PyAny, PyRef, PyResult, Python,
+    Bound, FromPyObject, Py, PyAny, PyObject, PyRef, PyResult, Python,
 };
 use pyo3_async_runtimes::tokio::future_into_py;

@@ -114,7 +114,7 @@ impl Connection {
        data: Bound<'_, PyAny>,
        namespace: Vec<String>,
        storage_options: Option<HashMap<String, String>>,
-        storage_options_provider: Option<Py<PyAny>>,
+        storage_options_provider: Option<PyObject>,
        location: Option<String>,
    ) -> PyResult<Bound<'a, PyAny>> {
        let inner = self_.get_inner()?.clone();
@@ -152,7 +152,7 @@ impl Connection {
        schema: Bound<'_, PyAny>,
        namespace: Vec<String>,
        storage_options: Option<HashMap<String, String>>,
-        storage_options_provider: Option<Py<PyAny>>,
+        storage_options_provider: Option<PyObject>,
        location: Option<String>,
    ) -> PyResult<Bound<'a, PyAny>> {
        let inner = self_.get_inner()?.clone();
@@ -187,7 +187,7 @@ impl Connection {
        name: String,
        namespace: Vec<String>,
        storage_options: Option<HashMap<String, String>>,
-        storage_options_provider: Option<Py<PyAny>>,
+        storage_options_provider: Option<PyObject>,
        index_cache_size: Option<u32>,
        location: Option<String>,
    ) -> PyResult<Bound<'_, PyAny>> {
@@ -304,10 +304,9 @@ impl Connection {
                },
                page_token,
                limit: limit.map(|l| l as i32),
-                ..Default::default()
            };
            let response = inner.list_namespaces(request).await.infer_error()?;
-            Python::attach(|py| -> PyResult<Py<PyDict>> {
+            Python::with_gil(|py| -> PyResult<Py<PyDict>> {
                let dict = PyDict::new(py);
                dict.set_item("namespaces", response.namespaces)?;
                dict.set_item("page_token", response.page_token)?;
@@ -326,12 +325,11 @@ impl Connection {
        let inner = self_.get_inner()?.clone();
        let py = self_.py();
        future_into_py(py, async move {
-            use lance_namespace::models::CreateNamespaceRequest;
-            // Mode is now a string field
-            let mode_str = mode.and_then(|m| match m.to_lowercase().as_str() {
-                "create" => Some("Create".to_string()),
-                "exist_ok" => Some("ExistOk".to_string()),
-                "overwrite" => Some("Overwrite".to_string()),
+            use lance_namespace::models::{create_namespace_request, CreateNamespaceRequest};
+            let mode_enum = mode.and_then(|m| match m.to_lowercase().as_str() {
+                "create" => Some(create_namespace_request::Mode::Create),
+                "exist_ok" => Some(create_namespace_request::Mode::ExistOk),
+                "overwrite" => Some(create_namespace_request::Mode::Overwrite),
                _ => None,
            });
            let request = CreateNamespaceRequest {
@@ -340,12 +338,11 @@ impl Connection {
                } else {
                    Some(namespace)
                },
-                mode: mode_str,
+                mode: mode_enum,
                properties,
-                ..Default::default()
            };
            let response = inner.create_namespace(request).await.infer_error()?;
-            Python::attach(|py| -> PyResult<Py<PyDict>> {
+            Python::with_gil(|py| -> PyResult<Py<PyDict>> {
                let dict = PyDict::new(py);
                dict.set_item("properties", response.properties)?;
                Ok(dict.unbind())
@@ -363,16 +360,15 @@ impl Connection {
        let inner = self_.get_inner()?.clone();
        let py = self_.py();
        future_into_py(py, async move {
-            use lance_namespace::models::DropNamespaceRequest;
-            // Mode and Behavior are now string fields
-            let mode_str = mode.and_then(|m| match m.to_uppercase().as_str() {
-                "SKIP" => Some("Skip".to_string()),
-                "FAIL" => Some("Fail".to_string()),
+            use lance_namespace::models::{drop_namespace_request, DropNamespaceRequest};
+            let mode_enum = mode.and_then(|m| match m.to_uppercase().as_str() {
+                "SKIP" => Some(drop_namespace_request::Mode::Skip),
+                "FAIL" => Some(drop_namespace_request::Mode::Fail),
                _ => None,
            });
-            let behavior_str = behavior.and_then(|b| match b.to_uppercase().as_str() {
-                "RESTRICT" => Some("Restrict".to_string()),
-                "CASCADE" => Some("Cascade".to_string()),
+            let behavior_enum = behavior.and_then(|b| match b.to_uppercase().as_str() {
+                "RESTRICT" => Some(drop_namespace_request::Behavior::Restrict),
+                "CASCADE" => Some(drop_namespace_request::Behavior::Cascade),
                _ => None,
            });
            let request = DropNamespaceRequest {
@@ -381,12 +377,11 @@ impl Connection {
                } else {
                    Some(namespace)
                },
-                mode: mode_str,
-                behavior: behavior_str,
-                ..Default::default()
+                mode: mode_enum,
+                behavior: behavior_enum,
            };
            let response = inner.drop_namespace(request).await.infer_error()?;
-            Python::attach(|py| -> PyResult<Py<PyDict>> {
+            Python::with_gil(|py| -> PyResult<Py<PyDict>> {
                let dict = PyDict::new(py);
                dict.set_item("properties", response.properties)?;
                dict.set_item("transaction_id", response.transaction_id)?;
@@ -410,10 +405,9 @@ impl Connection {
                } else {
                    Some(namespace)
                },
-                ..Default::default()
            };
            let response = inner.describe_namespace(request).await.infer_error()?;
-            Python::attach(|py| -> PyResult<Py<PyDict>> {
+            Python::with_gil(|py| -> PyResult<Py<PyDict>> {
                let dict = PyDict::new(py);
                dict.set_item("properties", response.properties)?;
                Ok(dict.unbind())
@@ -440,10 +434,9 @@ impl Connection {
                },
                page_token,
                limit: limit.map(|l| l as i32),
-                ..Default::default()
            };
            let response = inner.list_tables(request).await.infer_error()?;
-            Python::attach(|py| -> PyResult<Py<PyDict>> {
+            Python::with_gil(|py| -> PyResult<Py<PyDict>> {
                let dict = PyDict::new(py);
                dict.set_item("tables", response.tables)?;
                dict.set_item("page_token", response.page_token)?;
--- a/python/src/error.rs
+++ b/python/src/error.rs
@@ -40,7 +40,7 @@ impl<T> PythonErrorExt<T> for std::result::Result<T, LanceError> {
                    request_id,
                    source,
                    status_code,
-                } => Python::attach(|py| {
+                } => Python::with_gil(|py| {
                    let message = err.to_string();
                    let http_err_cls = py
                        .import(intern!(py, "lancedb.remote.errors"))?
@@ -75,7 +75,7 @@ impl<T> PythonErrorExt<T> for std::result::Result<T, LanceError> {
                    max_read_failures,
                    source,
                    status_code,
-                } => Python::attach(|py| {
+                } => Python::with_gil(|py| {
                    let cause_err = http_from_rust_error(
                        py,
                        source.as_ref(),
--- a/python/src/header.rs
+++ b/python/src/header.rs
@@ -12,7 +12,7 @@ pub struct PyHeaderProvider {

 impl Clone for PyHeaderProvider {
    fn clone(&self) -> Self {
-        Python::attach(|py| Self {
+        Python::with_gil(|py| Self {
            provider: self.provider.clone_ref(py),
        })
    }
@@ -25,7 +25,7 @@ impl PyHeaderProvider {

    /// Get headers from the Python provider (internal implementation)
    fn get_headers_internal(&self) -> Result<HashMap<String, String>, String> {
-        Python::attach(|py| {
+        Python::with_gil(|py| {
            // Call the get_headers method
            let result = self.provider.call_method0(py, "get_headers");

--- a/python/src/permutation.rs
+++ b/python/src/permutation.rs
@@ -281,7 +281,7 @@ impl PyPermutationReader {
        let reader = slf.reader.clone();
        future_into_py(slf.py(), async move {
            let schema = reader.output_schema(selection).await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

--- a/python/src/query.rs
+++ b/python/src/query.rs
@@ -453,7 +453,7 @@ impl Query {
        let inner = self_.inner.clone();
        future_into_py(self_.py(), async move {
            let schema = inner.output_schema().await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

@@ -532,7 +532,7 @@ impl TakeQuery {
        let inner = self_.inner.clone();
        future_into_py(self_.py(), async move {
            let schema = inner.output_schema().await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

@@ -627,7 +627,7 @@ impl FTSQuery {
        let inner = self_.inner.clone();
        future_into_py(self_.py(), async move {
            let schema = inner.output_schema().await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

@@ -806,7 +806,7 @@ impl VectorQuery {
        let inner = self_.inner.clone();
        future_into_py(self_.py(), async move {
            let schema = inner.output_schema().await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

--- a/python/src/storage_options.rs
+++ b/python/src/storage_options.rs
@@ -17,20 +17,20 @@ use pyo3::types::PyDict;
 /// Internal wrapper around a Python object implementing StorageOptionsProvider
 pub struct PyStorageOptionsProvider {
    /// The Python object implementing fetch_storage_options()
-    inner: Py<PyAny>,
+    inner: PyObject,
 }

 impl Clone for PyStorageOptionsProvider {
    fn clone(&self) -> Self {
-        Python::attach(|py| Self {
+        Python::with_gil(|py| Self {
            inner: self.inner.clone_ref(py),
        })
    }
 }

 impl PyStorageOptionsProvider {
-    pub fn new(obj: Py<PyAny>) -> PyResult<Self> {
-        Python::attach(|py| {
+    pub fn new(obj: PyObject) -> PyResult<Self> {
+        Python::with_gil(|py| {
            // Verify the object has a fetch_storage_options method
            if !obj.bind(py).hasattr("fetch_storage_options")? {
                return Err(pyo3::exceptions::PyTypeError::new_err(
@@ -60,7 +60,7 @@ impl StorageOptionsProvider for PyStorageOptionsProviderWrapper {
        let py_provider = self.py_provider.clone();

        tokio::task::spawn_blocking(move || {
-            Python::attach(|py| {
+            Python::with_gil(|py| {
                // Call the Python fetch_storage_options method
                let result = py_provider
                    .inner
@@ -119,7 +119,7 @@ impl StorageOptionsProvider for PyStorageOptionsProviderWrapper {
    }

    fn provider_id(&self) -> String {
-        Python::attach(|py| {
+        Python::with_gil(|py| {
            // Call provider_id() method on the Python object
            let obj = self.py_provider.inner.bind(py);
            obj.call_method0("provider_id")
@@ -143,7 +143,7 @@ impl std::fmt::Debug for PyStorageOptionsProviderWrapper {
 /// This is the main entry point for converting Python StorageOptionsProvider objects
 /// to Rust trait objects that can be used by the Lance ecosystem.
 pub fn py_object_to_storage_options_provider(
-    py_obj: Py<PyAny>,
+    py_obj: PyObject,
 ) -> PyResult<Arc<dyn StorageOptionsProvider>> {
    let py_provider = PyStorageOptionsProvider::new(py_obj)?;
    Ok(Arc::new(PyStorageOptionsProviderWrapper::new(py_provider)))
--- a/python/src/table.rs
+++ b/python/src/table.rs
@@ -287,7 +287,7 @@ impl Table {
        let inner = self_.inner_ref()?.clone();
        future_into_py(self_.py(), async move {
            let schema = inner.schema().await.infer_error()?;
-            Python::attach(|py| schema.to_pyarrow(py).map(|obj| obj.unbind()))
+            Python::with_gil(|py| schema.to_pyarrow(py))
        })
    }

@@ -437,7 +437,7 @@ impl Table {
        future_into_py(self_.py(), async move {
            let stats = inner.index_stats(&index_name).await.infer_error()?;
            if let Some(stats) = stats {
-                Python::attach(|py| {
+                Python::with_gil(|py| {
                    let dict = PyDict::new(py);
                    dict.set_item("num_indexed_rows", stats.num_indexed_rows)?;
                    dict.set_item("num_unindexed_rows", stats.num_unindexed_rows)?;
@@ -467,7 +467,7 @@ impl Table {
        let inner = self_.inner_ref()?.clone();
        future_into_py(self_.py(), async move {
            let stats = inner.stats().await.infer_error()?;
-            Python::attach(|py| {
+            Python::with_gil(|py| {
                let dict = PyDict::new(py);
                dict.set_item("total_bytes", stats.total_bytes)?;
                dict.set_item("num_rows", stats.num_rows)?;
@@ -497,11 +497,6 @@ impl Table {
        })
    }

-    pub fn uri(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self_.inner_ref()?.clone();
-        future_into_py(self_.py(), async move { inner.uri().await.infer_error() })
-    }
-
    pub fn __repr__(&self) -> String {
        match &self.inner {
            None => format!("ClosedTable({})", self.name),
@@ -521,7 +516,7 @@ impl Table {
        let inner = self_.inner_ref()?.clone();
        future_into_py(self_.py(), async move {
            let versions = inner.list_versions().await.infer_error()?;
-            let versions_as_dict = Python::attach(|py| {
+            let versions_as_dict = Python::with_gil(|py| {
                versions
                    .iter()
                    .map(|v| {
@@ -872,7 +867,7 @@ impl Tags {
            let tags = inner.tags().await.infer_error()?;
            let res = tags.list().await.infer_error()?;

-            Python::attach(|py| {
+            Python::with_gil(|py| {
                let py_dict = PyDict::new(py);
                for (key, contents) in res {
                    let value_dict = PyDict::new(py);
--- a/rust/lancedb/Cargo.toml
+++ b/rust/lancedb/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.24.0"
+version = "0.23.1"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true
@@ -104,16 +104,11 @@ test-log = "0.2"


 [features]
-default = []
+default = ["aws", "gcs", "azure", "dynamodb", "oss"]
 aws = ["lance/aws", "lance-io/aws", "lance-namespace-impls/dir-aws"]
 oss = ["lance/oss", "lance-io/oss", "lance-namespace-impls/dir-oss"]
 gcs = ["lance/gcp", "lance-io/gcp", "lance-namespace-impls/dir-gcp"]
 azure = ["lance/azure", "lance-io/azure", "lance-namespace-impls/dir-azure"]
-huggingface = [
-    "lance/huggingface",
-    "lance-io/huggingface",
-    "lance-namespace-impls/dir-huggingface",
-]
 dynamodb = ["lance/dynamodb", "aws"]
 remote = ["dep:reqwest", "dep:http", "lance-namespace-impls/rest", "lance-namespace-impls/rest-adapter"]
 fp16kernels = ["lance-linalg/fp16kernels"]
@@ -153,6 +148,3 @@ name = "ivf_pq"
 [[example]]
 name = "hybrid_search"
 required-features = ["sentence-transformers"]
-
-[package.metadata.docs.rs]
-all-features = true
--- a/rust/lancedb/src/connection.rs
+++ b/rust/lancedb/src/connection.rs
@@ -36,42 +36,10 @@ use crate::remote::{
 };
 use crate::table::{TableDefinition, WriteOptions};
 use crate::Table;
-use lance::io::ObjectStoreParams;
 pub use lance_encoding::version::LanceFileVersion;
 #[cfg(feature = "remote")]
 use lance_io::object_store::StorageOptions;
-use lance_io::object_store::{StorageOptionsAccessor, StorageOptionsProvider};
-
-fn merge_storage_options(
-    store_params: &mut ObjectStoreParams,
-    pairs: impl IntoIterator<Item = (String, String)>,
-) {
-    let mut options = store_params.storage_options().cloned().unwrap_or_default();
-    for (key, value) in pairs {
-        options.insert(key, value);
-    }
-    let provider = store_params
-        .storage_options_accessor
-        .as_ref()
-        .and_then(|accessor| accessor.provider().cloned());
-    let accessor = if let Some(provider) = provider {
-        StorageOptionsAccessor::with_initial_and_provider(options, provider)
-    } else {
-        StorageOptionsAccessor::with_static_options(options)
-    };
-    store_params.storage_options_accessor = Some(Arc::new(accessor));
-}
-
-fn set_storage_options_provider(
-    store_params: &mut ObjectStoreParams,
-    provider: Arc<dyn StorageOptionsProvider>,
-) {
-    let accessor = match store_params.storage_options().cloned() {
-        Some(options) => StorageOptionsAccessor::with_initial_and_provider(options, provider),
-        None => StorageOptionsAccessor::with_provider(provider),
-    };
-    store_params.storage_options_accessor = Some(Arc::new(accessor));
-}
+use lance_io::object_store::StorageOptionsProvider;

 /// A builder for configuring a [`Connection::table_names`] operation
 pub struct TableNamesBuilder {
@@ -278,14 +246,16 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
    ///
    /// See available options at <https://lancedb.com/docs/storage/>
    pub fn storage_option(mut self, key: impl Into<String>, value: impl Into<String>) -> Self {
-        let store_params = self
+        let store_options = self
            .request
            .write_options
            .lance_write_params
            .get_or_insert(Default::default())
            .store_params
+            .get_or_insert(Default::default())
+            .storage_options
            .get_or_insert(Default::default());
-        merge_storage_options(store_params, [(key.into(), value.into())]);
+        store_options.insert(key.into(), value.into());
        self
    }

@@ -299,17 +269,19 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
        mut self,
        pairs: impl IntoIterator<Item = (impl Into<String>, impl Into<String>)>,
    ) -> Self {
-        let store_params = self
+        let store_options = self
            .request
            .write_options
            .lance_write_params
            .get_or_insert(Default::default())
            .store_params
+            .get_or_insert(Default::default())
+            .storage_options
            .get_or_insert(Default::default());
-        let updates = pairs
-            .into_iter()
-            .map(|(key, value)| (key.into(), value.into()));
-        merge_storage_options(store_params, updates);
+
+        for (key, value) in pairs {
+            store_options.insert(key.into(), value.into());
+        }
        self
    }

@@ -346,21 +318,23 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
    /// This has no effect in LanceDB Cloud.
    #[deprecated(since = "0.15.1", note = "Use `database_options` instead")]
    pub fn enable_v2_manifest_paths(mut self, use_v2_manifest_paths: bool) -> Self {
-        let store_params = self
+        let storage_options = self
            .request
            .write_options
            .lance_write_params
            .get_or_insert_with(Default::default)
            .store_params
+            .get_or_insert_with(Default::default)
+            .storage_options
            .get_or_insert_with(Default::default);
-        let value = if use_v2_manifest_paths {
-            "true".to_string()
-        } else {
-            "false".to_string()
-        };
-        merge_storage_options(
-            store_params,
-            [(OPT_NEW_TABLE_V2_MANIFEST_PATHS.to_string(), value)],
+
+        storage_options.insert(
+            OPT_NEW_TABLE_V2_MANIFEST_PATHS.to_string(),
+            if use_v2_manifest_paths {
+                "true".to_string()
+            } else {
+                "false".to_string()
+            },
        );
        self
    }
@@ -370,19 +344,19 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
    /// The default is `LanceFileVersion::Stable`.
    #[deprecated(since = "0.15.1", note = "Use `database_options` instead")]
    pub fn data_storage_version(mut self, data_storage_version: LanceFileVersion) -> Self {
-        let store_params = self
+        let storage_options = self
            .request
            .write_options
            .lance_write_params
            .get_or_insert_with(Default::default)
            .store_params
+            .get_or_insert_with(Default::default)
+            .storage_options
            .get_or_insert_with(Default::default);
-        merge_storage_options(
-            store_params,
-            [(
-                OPT_NEW_TABLE_STORAGE_VERSION.to_string(),
-                data_storage_version.to_string(),
-            )],
+
+        storage_options.insert(
+            OPT_NEW_TABLE_STORAGE_VERSION.to_string(),
+            data_storage_version.to_string(),
        );
        self
    }
@@ -407,14 +381,13 @@ impl<const HAS_DATA: bool> CreateTableBuilder<HAS_DATA> {
    /// This allows tables to automatically refresh cloud storage credentials
    /// when they expire, enabling long-running operations on remote storage.
    pub fn storage_options_provider(mut self, provider: Arc<dyn StorageOptionsProvider>) -> Self {
-        let store_params = self
-            .request
+        self.request
            .write_options
            .lance_write_params
            .get_or_insert(Default::default())
            .store_params
-            .get_or_insert(Default::default());
-        set_storage_options_provider(store_params, provider);
+            .get_or_insert(Default::default())
+            .storage_options_provider = Some(provider);
        self
    }
 }
@@ -477,13 +450,15 @@ impl OpenTableBuilder {
    ///
    /// See available options at <https://lancedb.com/docs/storage/>
    pub fn storage_option(mut self, key: impl Into<String>, value: impl Into<String>) -> Self {
-        let store_params = self
+        let storage_options = self
            .request
            .lance_read_params
            .get_or_insert(Default::default())
            .store_options
+            .get_or_insert(Default::default())
+            .storage_options
            .get_or_insert(Default::default());
-        merge_storage_options(store_params, [(key.into(), value.into())]);
+        storage_options.insert(key.into(), value.into());
        self
    }

@@ -497,16 +472,18 @@ impl OpenTableBuilder {
        mut self,
        pairs: impl IntoIterator<Item = (impl Into<String>, impl Into<String>)>,
    ) -> Self {
-        let store_params = self
+        let storage_options = self
            .request
            .lance_read_params
            .get_or_insert(Default::default())
            .store_options
+            .get_or_insert(Default::default())
+            .storage_options
            .get_or_insert(Default::default());
-        let updates = pairs
-            .into_iter()
-            .map(|(key, value)| (key.into(), value.into()));
-        merge_storage_options(store_params, updates);
+
+        for (key, value) in pairs {
+            storage_options.insert(key.into(), value.into());
+        }
        self
    }

@@ -530,13 +507,12 @@ impl OpenTableBuilder {
    /// This allows tables to automatically refresh cloud storage credentials
    /// when they expire, enabling long-running operations on remote storage.
    pub fn storage_options_provider(mut self, provider: Arc<dyn StorageOptionsProvider>) -> Self {
-        let store_params = self
-            .request
+        self.request
            .lance_read_params
            .get_or_insert(Default::default())
            .store_options
-            .get_or_insert(Default::default());
-        set_storage_options_provider(store_params, provider);
+            .get_or_insert(Default::default())
+            .storage_options_provider = Some(provider);
        self
    }

@@ -892,9 +868,6 @@ pub struct ConnectBuilder {
    embedding_registry: Option<Arc<dyn EmbeddingRegistry>>,
 }

-const ENV_VARS_TO_STORAGE_OPTS: [(&str, &str); 1] =
-    [("AZURE_STORAGE_ACCOUNT_NAME", "azure_storage_account_name")];
-
 impl ConnectBuilder {
    /// Create a new [`ConnectOptions`] with the given database URI.
    pub fn new(uri: &str) -> Self {
@@ -1078,27 +1051,11 @@ impl ConnectBuilder {
        self
    }

-    #[cfg(feature = "remote")]
-    fn apply_env_defaults(
-        env_var_to_remote_storage_option: &[(&str, &str)],
-        options: &mut HashMap<String, String>,
-    ) {
-        for (env_key, opt_key) in env_var_to_remote_storage_option {
-            if let Ok(env_value) = std::env::var(env_key) {
-                if !options.contains_key(*opt_key) {
-                    options.insert((*opt_key).to_string(), env_value);
-                }
-            }
-        }
-    }
-
    #[cfg(feature = "remote")]
    fn execute_remote(self) -> Result<Connection> {
        use crate::remote::db::RemoteDatabaseOptions;

-        let mut merged_options = self.request.options.clone();
-        Self::apply_env_defaults(&ENV_VARS_TO_STORAGE_OPTS, &mut merged_options);
-        let options = RemoteDatabaseOptions::parse_from_map(&merged_options)?;
+        let options = RemoteDatabaseOptions::parse_from_map(&self.request.options)?;

        let region = options.region.ok_or_else(|| Error::InvalidInput {
            message: "A region is required when connecting to LanceDb Cloud".to_string(),
@@ -1320,6 +1277,8 @@ mod test_utils {

 #[cfg(test)]
 mod tests {
+    use std::fs::create_dir_all;
+
    use crate::database::listing::{ListingDatabaseOptions, NewTableConfig};
    use crate::query::QueryBase;
    use crate::query::{ExecutableQuery, QueryExecutionOptions};
@@ -1343,23 +1302,6 @@ mod tests {
        assert_eq!(tc.connection.uri(), tc.uri);
    }

-    #[cfg(feature = "remote")]
-    #[test]
-    fn test_apply_env_defaults() {
-        let env_key = "TEST_APPLY_ENV_DEFAULTS_ENVIRONMENT_VARIABLE_ENV_KEY";
-        let env_val = "TEST_APPLY_ENV_DEFAULTS_ENVIRONMENT_VARIABLE_ENV_VAL";
-        let opts_key = "test_apply_env_defaults_environment_variable_opts_key";
-        std::env::set_var(env_key, env_val);
-
-        let mut options = HashMap::new();
-        ConnectBuilder::apply_env_defaults(&[(env_key, opts_key)], &mut options);
-        assert_eq!(Some(&env_val.to_string()), options.get(opts_key));
-
-        options.insert(opts_key.to_string(), "EXPLICIT-VALUE".to_string());
-        ConnectBuilder::apply_env_defaults(&[(env_key, opts_key)], &mut options);
-        assert_eq!(Some(&"EXPLICIT-VALUE".to_string()), options.get(opts_key));
-    }
-
    #[cfg(not(windows))]
    #[tokio::test]
    async fn test_connect_relative() {
@@ -1383,27 +1325,25 @@ mod tests {

    #[tokio::test]
    async fn test_table_names() {
-        let tc = new_test_connection().await.unwrap();
-        let db = tc.connection;
-        let schema = Arc::new(Schema::new(vec![Field::new("x", DataType::Int32, false)]));
+        let tmp_dir = tempdir().unwrap();
        let mut names = Vec::with_capacity(100);
        for _ in 0..100 {
-            let name = uuid::Uuid::new_v4().to_string();
+            let mut name = uuid::Uuid::new_v4().to_string();
            names.push(name.clone());
-            db.create_empty_table(name, schema.clone())
-                .execute()
-                .await
-                .unwrap();
+            name.push_str(".lance");
+            create_dir_all(tmp_dir.path().join(&name)).unwrap();
        }
        names.sort();
-        let tables = db.table_names().limit(100).execute().await.unwrap();
+
+        let uri = tmp_dir.path().to_str().unwrap();
+        let db = connect(uri).execute().await.unwrap();
+        let tables = db.table_names().execute().await.unwrap();

        assert_eq!(tables, names);

        let tables = db
            .table_names()
            .start_after(&names[30])
-            .limit(100)
            .execute()
            .await
            .unwrap();
@@ -1584,27 +1524,18 @@ mod tests {

    #[tokio::test]
    async fn drop_table() {
-        let tc = new_test_connection().await.unwrap();
-        let db = tc.connection;
+        let tmp_dir = tempdir().unwrap();

-        if tc.is_remote {
-            // All the typical endpoints such as s3:///, file-object-store:///, etc. treat drop_table
-            // as idempotent.
-            assert!(db.drop_table("invalid_table", &[]).await.is_ok());
-        } else {
-            // The behavior of drop_table when using a file:/// endpoint differs from all other
-            // object providers, in that it returns an error when deleting a non-existent table.
-            assert!(matches!(
-                db.drop_table("invalid_table", &[]).await,
-                Err(crate::Error::TableNotFound { .. }),
-            ));
-        }
+        let uri = tmp_dir.path().to_str().unwrap();
+        let db = connect(uri).execute().await.unwrap();

-        let schema = Arc::new(Schema::new(vec![Field::new("x", DataType::Int32, false)]));
-        db.create_empty_table("table1", schema.clone())
-            .execute()
-            .await
-            .unwrap();
+        // drop non-exist table
+        assert!(matches!(
+            db.drop_table("invalid_table", &[]).await,
+            Err(crate::Error::TableNotFound { .. }),
+        ));
+
+        create_dir_all(tmp_dir.path().join("table1.lance")).unwrap();
        db.drop_table("table1", &[]).await.unwrap();

        let tables = db.table_names().execute().await.unwrap();
--- a/rust/lancedb/src/database/listing.rs
+++ b/rust/lancedb/src/database/listing.rs
@@ -12,7 +12,7 @@ use lance::dataset::{builder::DatasetBuilder, ReadParams, WriteMode};
 use lance::io::{ObjectStore, ObjectStoreParams, WrappingObjectStore};
 use lance_datafusion::utils::StreamingWriteSource;
 use lance_encoding::version::LanceFileVersion;
-use lance_io::object_store::{StorageOptionsAccessor, StorageOptionsProvider};
+use lance_io::object_store::StorageOptionsProvider;
 use lance_table::io::commit::commit_handler_from_url;
 use object_store::local::LocalFileSystem;
 use snafu::ResultExt;
@@ -356,13 +356,7 @@ impl ListingDatabase {
                    .clone()
                    .unwrap_or_else(|| Arc::new(lance::session::Session::default()));
                let os_params = ObjectStoreParams {
-                    storage_options_accessor: if options.storage_options.is_empty() {
-                        None
-                    } else {
-                        Some(Arc::new(StorageOptionsAccessor::with_static_options(
-                            options.storage_options.clone(),
-                        )))
-                    },
+                    storage_options: Some(options.storage_options.clone()),
                    ..Default::default()
                };
                let (object_store, base_path) = ObjectStore::from_uri_and_params(
@@ -469,20 +463,9 @@ impl ListingDatabase {
        validate_table_name(name)?;

        let mut uri = self.uri.clone();
-        // If the URI does not end with a path separator, add one
-        // Use forward slash for URIs (http://, s3://, gs://, file://, etc.)
-        // Use platform-specific separator for local paths without scheme
-        let has_scheme = uri.contains("://");
-        let ends_with_separator = uri.ends_with('/') || uri.ends_with('\\');
-
-        if !ends_with_separator {
-            if has_scheme {
-                // URIs always use forward slash
-                uri.push('/');
-            } else {
-                // Local path without scheme - use platform separator
-                uri.push(std::path::MAIN_SEPARATOR);
-            }
+        // If the URI does not end with a slash, add one
+        if !uri.ends_with('/') {
+            uri.push('/');
        }
        // Append the table name with the lance file extension
        uri.push_str(&format!("{}.{}", name, LANCE_FILE_EXTENSION));
@@ -498,13 +481,7 @@ impl ListingDatabase {

    async fn drop_tables(&self, names: Vec<String>) -> Result<()> {
        let object_store_params = ObjectStoreParams {
-            storage_options_accessor: if self.storage_options.is_empty() {
-                None
-            } else {
-                Some(Arc::new(StorageOptionsAccessor::with_static_options(
-                    self.storage_options.clone(),
-                )))
-            },
+            storage_options: Some(self.storage_options.clone()),
            ..Default::default()
        };
        let mut uri = self.uri.clone();
@@ -553,7 +530,7 @@ impl ListingDatabase {
            .lance_write_params
            .as_ref()
            .and_then(|p| p.store_params.as_ref())
-            .and_then(|sp| sp.storage_options());
+            .and_then(|sp| sp.storage_options.as_ref());

        let storage_version_override = storage_options
            .and_then(|opts| opts.get(OPT_NEW_TABLE_STORAGE_VERSION))
@@ -604,20 +581,21 @@ impl ListingDatabase {
        // will cause a new connection to be created, and that connection will
        // be dropped from the cache when python GCs the table object, which
        // confounds reuse across tables.
-        if !self.storage_options.is_empty() || self.storage_options_provider.is_some() {
-            let store_params = write_params
+        if !self.storage_options.is_empty() {
+            let storage_options = write_params
                .store_params
+                .get_or_insert_with(Default::default)
+                .storage_options
                .get_or_insert_with(Default::default);
-            let mut storage_options = store_params.storage_options().cloned().unwrap_or_default();
-            if !self.storage_options.is_empty() {
-                self.inherit_storage_options(&mut storage_options);
-            }
-            let accessor = if let Some(ref provider) = self.storage_options_provider {
-                StorageOptionsAccessor::with_initial_and_provider(storage_options, provider.clone())
-            } else {
-                StorageOptionsAccessor::with_static_options(storage_options)
-            };
-            store_params.storage_options_accessor = Some(Arc::new(accessor));
+            self.inherit_storage_options(storage_options);
+        }
+
+        // Set storage options provider if available
+        if self.storage_options_provider.is_some() {
+            write_params
+                .store_params
+                .get_or_insert_with(Default::default)
+                .storage_options_provider = self.storage_options_provider.clone();
        }

        write_params.data_storage_version = self
@@ -903,13 +881,7 @@ impl Database for ListingDatabase {
        validate_table_name(&request.target_table_name)?;

        let storage_params = ObjectStoreParams {
-            storage_options_accessor: if self.storage_options.is_empty() {
-                None
-            } else {
-                Some(Arc::new(StorageOptionsAccessor::with_static_options(
-                    self.storage_options.clone(),
-                )))
-            },
+            storage_options: Some(self.storage_options.clone()),
            ..Default::default()
        };
        let read_params = ReadParams {
@@ -973,28 +945,25 @@ impl Database for ListingDatabase {
        // will cause a new connection to be created, and that connection will
        // be dropped from the cache when python GCs the table object, which
        // confounds reuse across tables.
-        if !self.storage_options.is_empty() || self.storage_options_provider.is_some() {
-            let store_params = request
+        if !self.storage_options.is_empty() {
+            let storage_options = request
                .lance_read_params
                .get_or_insert_with(Default::default)
                .store_options
+                .get_or_insert_with(Default::default)
+                .storage_options
                .get_or_insert_with(Default::default);
-            let mut storage_options = store_params.storage_options().cloned().unwrap_or_default();
-            if !self.storage_options.is_empty() {
-                self.inherit_storage_options(&mut storage_options);
-            }
-            // Preserve request-level provider if no connection-level provider exists
-            let request_provider = store_params
-                .storage_options_accessor
-                .as_ref()
-                .and_then(|a| a.provider().cloned());
-            let provider = self.storage_options_provider.clone().or(request_provider);
-            let accessor = if let Some(provider) = provider {
-                StorageOptionsAccessor::with_initial_and_provider(storage_options, provider)
-            } else {
-                StorageOptionsAccessor::with_static_options(storage_options)
-            };
-            store_params.storage_options_accessor = Some(Arc::new(accessor));
+            self.inherit_storage_options(storage_options);
+        }
+
+        // Set storage options provider if available
+        if self.storage_options_provider.is_some() {
+            request
+                .lance_read_params
+                .get_or_insert_with(Default::default)
+                .store_options
+                .get_or_insert_with(Default::default)
+                .storage_options_provider = self.storage_options_provider.clone();
        }

        // Some ReadParams are exposed in the OpenTableBuilder, but we also
@@ -1102,7 +1071,6 @@ mod tests {
    use crate::table::{Table, TableDefinition};
    use arrow_array::{Int32Array, RecordBatch, StringArray};
    use arrow_schema::{DataType, Field, Schema};
-    use std::path::PathBuf;
    use tempfile::tempdir;

    async fn setup_database() -> (tempfile::TempDir, ListingDatabase) {
@@ -1901,9 +1869,7 @@ mod tests {
        let write_options = WriteOptions {
            lance_write_params: Some(lance::dataset::WriteParams {
                store_params: Some(lance::io::ObjectStoreParams {
-                    storage_options_accessor: Some(Arc::new(
-                        StorageOptionsAccessor::with_static_options(storage_options),
-                    )),
+                    storage_options: Some(storage_options),
                    ..Default::default()
                }),
                ..Default::default()
@@ -1977,9 +1943,7 @@ mod tests {
        let write_options = WriteOptions {
            lance_write_params: Some(lance::dataset::WriteParams {
                store_params: Some(lance::io::ObjectStoreParams {
-                    storage_options_accessor: Some(Arc::new(
-                        StorageOptionsAccessor::with_static_options(storage_options),
-                    )),
+                    storage_options: Some(storage_options),
                    ..Default::default()
                }),
                ..Default::default()
@@ -2082,19 +2046,6 @@ mod tests {
        assert_eq!(db_options.new_table_config.enable_stable_row_ids, None);
    }

-    #[tokio::test]
-    async fn test_table_uri() {
-        let (_tempdir, db) = setup_database().await;
-
-        let mut pb = PathBuf::new();
-        pb.push(db.uri.clone());
-        pb.push("test.lance");
-
-        let expected = pb.to_str().unwrap();
-        let uri = db.table_uri("test").ok().unwrap();
-        assert_eq!(uri, expected);
-    }
-
    #[tokio::test]
    async fn test_namespace_client() {
        let (_tempdir, db) = setup_database().await;
--- a/rust/lancedb/src/database/namespace.rs
+++ b/rust/lancedb/src/database/namespace.rs
@@ -10,14 +10,13 @@ use async_trait::async_trait;
 use lance_namespace::{
    models::{
        CreateEmptyTableRequest, CreateNamespaceRequest, CreateNamespaceResponse,
-        DeclareTableRequest, DescribeNamespaceRequest, DescribeNamespaceResponse,
-        DescribeTableRequest, DropNamespaceRequest, DropNamespaceResponse, DropTableRequest,
-        ListNamespacesRequest, ListNamespacesResponse, ListTablesRequest, ListTablesResponse,
+        DescribeNamespaceRequest, DescribeNamespaceResponse, DescribeTableRequest,
+        DropNamespaceRequest, DropNamespaceResponse, DropTableRequest, ListNamespacesRequest,
+        ListNamespacesResponse, ListTablesRequest, ListTablesResponse,
    },
    LanceNamespace,
 };
 use lance_namespace_impls::ConnectBuilder;
-use log::warn;

 use crate::database::ReadConsistency;
 use crate::error::{Error, Result};
@@ -138,7 +137,6 @@ impl Database for LanceNamespaceDatabase {
            id: Some(request.namespace),
            page_token: request.start_after,
            limit: request.limit.map(|l| l as i32),
-            ..Default::default()
        };

        let response = self.namespace.list_tables(ns_request).await?;
@@ -155,7 +153,7 @@ impl Database for LanceNamespaceDatabase {
        table_id.push(request.name.clone());
        let describe_request = DescribeTableRequest {
            id: Some(table_id.clone()),
-            ..Default::default()
+            version: None,
        };

        let describe_result = self.namespace.describe_table(describe_request).await;
@@ -173,7 +171,6 @@ impl Database for LanceNamespaceDatabase {
                    // Drop the existing table - must succeed
                    let drop_request = DropTableRequest {
                        id: Some(table_id.clone()),
-                        ..Default::default()
                    };
                    self.namespace
                        .drop_table(drop_request)
@@ -205,53 +202,29 @@ impl Database for LanceNamespaceDatabase {
        let mut table_id = request.namespace.clone();
        table_id.push(request.name.clone());

-        // Try declare_table first, falling back to create_empty_table for backwards
-        // compatibility with older namespace clients that don't support declare_table
-        let declare_request = DeclareTableRequest {
+        let create_empty_request = CreateEmptyTableRequest {
            id: Some(table_id.clone()),
-            ..Default::default()
+            location: None,
+            properties: if self.storage_options.is_empty() {
+                None
+            } else {
+                Some(self.storage_options.clone())
+            },
        };

-        let location = match self.namespace.declare_table(declare_request).await {
-            Ok(response) => response.location.ok_or_else(|| Error::Runtime {
-                message: "Table location is missing from declare_table response".to_string(),
-            })?,
-            Err(e) => {
-                // Check if the error is "not supported" and try create_empty_table as fallback
-                let err_str = e.to_string().to_lowercase();
-                if err_str.contains("not supported") || err_str.contains("not implemented") {
-                    warn!(
-                        "declare_table is not supported by the namespace client, \
-                        falling back to deprecated create_empty_table. \
-                        create_empty_table is deprecated and will be removed in Lance 3.0.0. \
-                        Please upgrade your namespace client to support declare_table."
-                    );
-                    #[allow(deprecated)]
-                    let create_empty_request = CreateEmptyTableRequest {
-                        id: Some(table_id.clone()),
-                        ..Default::default()
-                    };
+        let create_empty_response = self
+            .namespace
+            .create_empty_table(create_empty_request)
+            .await
+            .map_err(|e| Error::Runtime {
+                message: format!("Failed to create empty table: {}", e),
+            })?;

-                    #[allow(deprecated)]
-                    let create_response = self
-                        .namespace
-                        .create_empty_table(create_empty_request)
-                        .await
-                        .map_err(|e| Error::Runtime {
-                            message: format!("Failed to create empty table: {}", e),
-                        })?;
-
-                    create_response.location.ok_or_else(|| Error::Runtime {
-                        message: "Table location is missing from create_empty_table response"
-                            .to_string(),
-                    })?
-                } else {
-                    return Err(Error::Runtime {
-                        message: format!("Failed to declare table: {}", e),
-                    });
-                }
-            }
-        };
+        let location = create_empty_response
+            .location
+            .ok_or_else(|| Error::Runtime {
+                message: "Table location is missing from create_empty_table response".to_string(),
+            })?;

        let native_table = NativeTable::create_from_namespace(
            self.namespace.clone(),
@@ -308,10 +281,7 @@ impl Database for LanceNamespaceDatabase {
        let mut table_id = namespace.to_vec();
        table_id.push(name.to_string());

-        let drop_request = DropTableRequest {
-            id: Some(table_id),
-            ..Default::default()
-        };
+        let drop_request = DropTableRequest { id: Some(table_id) };
        self.namespace
            .drop_table(drop_request)
            .await
@@ -466,7 +436,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -526,7 +497,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -589,7 +561,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -672,7 +645,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -727,7 +701,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -807,7 +782,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
@@ -840,7 +816,8 @@ mod tests {
        // Create a child namespace first
        conn.create_namespace(CreateNamespaceRequest {
            id: Some(vec!["test_ns".into()]),
-            ..Default::default()
+            mode: None,
+            properties: None,
        })
        .await
        .expect("Failed to create namespace");
--- a/rust/lancedb/src/embeddings.rs
+++ b/rust/lancedb/src/embeddings.rs
@@ -120,13 +120,8 @@ impl MemoryRegistry {
 }

 /// A record batch reader that has embeddings applied to it
-///
-/// This is a wrapper around another record batch reader that applies embedding functions
-/// when reading from the record batch.
-///
-/// When multiple embedding functions are defined, they are computed in parallel using
-/// scoped threads to improve performance. For a single embedding function, computation
-/// is done inline without threading overhead.
+/// This is a wrapper around another record batch reader that applies an embedding function
+/// when reading from the record batch
 pub struct WithEmbeddings<R: RecordBatchReader> {
    inner: R,
    embeddings: Vec<(EmbeddingDefinition, Arc<dyn EmbeddingFunction>)>,
@@ -240,48 +235,6 @@ impl<R: RecordBatchReader> WithEmbeddings<R> {
            column_definitions,
        })
    }
-
-    fn compute_embeddings_parallel(&self, batch: &RecordBatch) -> Result<Vec<Arc<dyn Array>>> {
-        if self.embeddings.len() == 1 {
-            let (fld, func) = &self.embeddings[0];
-            let src_column =
-                batch
-                    .column_by_name(&fld.source_column)
-                    .ok_or_else(|| Error::InvalidInput {
-                        message: format!("Source column '{}' not found", fld.source_column),
-                    })?;
-            return Ok(vec![func.compute_source_embeddings(src_column.clone())?]);
-        }
-
-        // Parallel path: multiple embeddings
-        std::thread::scope(|s| {
-            let handles: Vec<_> = self
-                .embeddings
-                .iter()
-                .map(|(fld, func)| {
-                    let src_column = batch.column_by_name(&fld.source_column).ok_or_else(|| {
-                        Error::InvalidInput {
-                            message: format!("Source column '{}' not found", fld.source_column),
-                        }
-                    })?;
-
-                    let handle =
-                        s.spawn(move || func.compute_source_embeddings(src_column.clone()));
-
-                    Ok(handle)
-                })
-                .collect::<Result<_>>()?;
-
-            handles
-                .into_iter()
-                .map(|h| {
-                    h.join().map_err(|e| Error::Runtime {
-                        message: format!("Thread panicked during embedding computation: {:?}", e),
-                    })?
-                })
-                .collect()
-        })
-    }
 }

 impl<R: RecordBatchReader> Iterator for MaybeEmbedded<R> {
@@ -309,19 +262,19 @@ impl<R: RecordBatchReader> Iterator for WithEmbeddings<R> {
    fn next(&mut self) -> Option<Self::Item> {
        let batch = self.inner.next()?;
        match batch {
-            Ok(batch) => {
-                let embeddings = match self.compute_embeddings_parallel(&batch) {
-                    Ok(emb) => emb,
-                    Err(e) => {
-                        return Some(Err(arrow_schema::ArrowError::ComputeError(format!(
-                            "Error computing embedding: {}",
-                            e
-                        ))))
-                    }
-                };
-
-                let mut batch = batch;
-                for ((fld, _), embedding) in self.embeddings.iter().zip(embeddings.iter()) {
+            Ok(mut batch) => {
+                // todo: parallelize this
+                for (fld, func) in self.embeddings.iter() {
+                    let src_column = batch.column_by_name(&fld.source_column).unwrap();
+                    let embedding = match func.compute_source_embeddings(src_column.clone()) {
+                        Ok(embedding) => embedding,
+                        Err(e) => {
+                            return Some(Err(arrow_schema::ArrowError::ComputeError(format!(
+                                "Error computing embedding: {}",
+                                e
+                            ))))
+                        }
+                    };
                    let dst_field_name = fld
                        .dest_column
                        .clone()
@@ -333,7 +286,7 @@ impl<R: RecordBatchReader> Iterator for WithEmbeddings<R> {
                        embedding.nulls().is_some(),
                    );

-                    match batch.try_with_column(dst_field.clone(), embedding.clone()) {
+                    match batch.try_with_column(dst_field.clone(), embedding) {
                        Ok(b) => batch = b,
                        Err(e) => return Some(Err(e)),
                    };
--- a/rust/lancedb/src/lib.rs
+++ b/rust/lancedb/src/lib.rs
@@ -25,14 +25,13 @@
 //!
 //! ## Crate Features
 //!
-//! - `aws` - Enable AWS S3 object store support.
-//! - `dynamodb` - Enable DynamoDB manifest store support.
-//! - `azure` - Enable Azure Blob Storage object store support.
-//! - `gcs` - Enable Google Cloud Storage object store support.
-//! - `oss` - Enable Alibaba Cloud OSS object store support.
-//! - `remote` - Enable remote client to connect to LanceDB cloud.
-//! - `huggingface` - Enable HuggingFace Hub integration for loading datasets from the Hub.
-//! - `fp16kernels` - Enable FP16 kernels for faster vector search on CPU.
+//! ### Experimental Features
+//!
+//! These features are not enabled by default.  They are experimental or in-development features that
+//! are not yet ready to be released.
+//!
+//! - `remote` - Enable remote client to connect to LanceDB cloud.  This is not yet fully implemented
+//!   and should not be enabled.
 //!
 //! ### Quick Start
 //!
@@ -51,15 +50,17 @@
 //! - `s3://bucket/path/to/database` or `gs://bucket/path/to/database` - database on cloud object store
 //! - `db://dbname` - Lance Cloud
 //!
-//! You can also use [`ConnectBuilder`] to configure the connection to the database.
+//! You can also use [`ConnectOptions`] to configure the connection to the database.
 //!
 //! ```rust
+//! use object_store::aws::AwsCredential;
 //! # tokio::runtime::Runtime::new().unwrap().block_on(async {
 //! let db = lancedb::connect("data/sample-lancedb")
-//!     .storage_options([
-//!         ("aws_access_key_id", "some_key"),
-//!         ("aws_secret_access_key", "some_secret"),
-//!     ])
+//!     .aws_creds(AwsCredential {
+//!         key_id: "some_key".to_string(),
+//!         secret_key: "some_secret".to_string(),
+//!         token: None,
+//!     })
 //!     .execute()
 //!     .await
 //!     .unwrap();
--- a/rust/lancedb/src/remote/db.rs
+++ b/rust/lancedb/src/remote/db.rs
@@ -1718,7 +1718,8 @@ mod tests {
            let namespace = vec!["test_ns".to_string()];
            conn.create_namespace(CreateNamespaceRequest {
                id: Some(namespace.clone()),
-                ..Default::default()
+                mode: None,
+                properties: None,
            })
            .await
            .expect("Failed to create namespace");
@@ -1743,7 +1744,8 @@ mod tests {
            let list_response = conn
                .list_tables(ListTablesRequest {
                    id: Some(namespace.clone()),
-                    ..Default::default()
+                    page_token: None,
+                    limit: None,
                })
                .await
                .expect("Failed to list tables");
@@ -1754,7 +1756,8 @@ mod tests {
            let list_response = namespace_client
                .list_tables(ListTablesRequest {
                    id: Some(namespace.clone()),
-                    ..Default::default()
+                    page_token: None,
+                    limit: None,
                })
                .await
                .unwrap();
@@ -1794,7 +1797,8 @@ mod tests {
            let namespace = vec!["multi_table_ns".to_string()];
            conn.create_namespace(CreateNamespaceRequest {
                id: Some(namespace.clone()),
-                ..Default::default()
+                mode: None,
+                properties: None,
            })
            .await
            .expect("Failed to create namespace");
@@ -1819,7 +1823,8 @@ mod tests {
            let list_response = conn
                .list_tables(ListTablesRequest {
                    id: Some(namespace.clone()),
-                    ..Default::default()
+                    page_token: None,
+                    limit: None,
                })
                .await
                .unwrap();
--- a/rust/lancedb/src/remote/table.rs
+++ b/rust/lancedb/src/remote/table.rs
@@ -204,7 +204,6 @@ pub struct RemoteTable<S: HttpSend = Sender> {
    server_version: ServerVersion,

    version: RwLock<Option<u64>>,
-    location: RwLock<Option<String>>,
 }

 impl<S: HttpSend> RemoteTable<S> {
@@ -222,7 +221,6 @@ impl<S: HttpSend> RemoteTable<S> {
            identifier,
            server_version,
            version: RwLock::new(None),
-            location: RwLock::new(None),
        }
    }

@@ -468,9 +466,7 @@ impl<S: HttpSend> RemoteTable<S> {
        self.apply_query_params(&mut body, &query.base)?;

        // Apply general parameters, before we dispatch based on number of query vectors.
-        if let Some(distance_type) = query.distance_type {
-            body["distance_type"] = serde_json::json!(distance_type);
-        }
+        body["distance_type"] = serde_json::json!(query.distance_type.unwrap_or_default());
        // In 0.23.1 we migrated from `nprobes` to `minimum_nprobes` and `maximum_nprobes`.
        // Old client / new server: since minimum_nprobes is missing, fallback to nprobes
        // New client / old server: old server will only see nprobes, make sure to set both
@@ -643,7 +639,6 @@ impl<S: HttpSend> RemoteTable<S> {
 struct TableDescription {
    version: u64,
    schema: JsonSchema,
-    location: Option<String>,
 }

 impl<S: HttpSend> std::fmt::Display for RemoteTable<S> {
@@ -672,7 +667,6 @@ mod test_utils {
                identifier: name,
                server_version: version.map(ServerVersion).unwrap_or_default(),
                version: RwLock::new(None),
-                location: RwLock::new(None),
            }
        }
    }
@@ -1094,17 +1088,6 @@ impl<S: HttpSend> BaseTable for RemoteTable<S> {
                    body["num_partitions"] = serde_json::Value::Number(num_partitions.into());
                }
            }
-            Index::IvfRq(index) => {
-                body[INDEX_TYPE_KEY] = serde_json::Value::String("IVF_RQ".to_string());
-                body[METRIC_TYPE_KEY] =
-                    serde_json::Value::String(index.distance_type.to_string().to_lowercase());
-                if let Some(num_partitions) = index.num_partitions {
-                    body["num_partitions"] = serde_json::Value::Number(num_partitions.into());
-                }
-                if let Some(num_bits) = index.num_bits {
-                    body["num_bits"] = serde_json::Value::Number(num_bits.into());
-                }
-            }
            Index::BTree(_) => {
                body[INDEX_TYPE_KEY] = serde_json::Value::String("BTREE".to_string());
            }
@@ -1467,28 +1450,8 @@ impl<S: HttpSend> BaseTable for RemoteTable<S> {
            message: "table_definition is not supported on LanceDB cloud.".into(),
        })
    }
-    async fn uri(&self) -> Result<String> {
-        // Check if we already have the location cached
-        {
-            let location = self.location.read().await;
-            if let Some(ref loc) = *location {
-                return Ok(loc.clone());
-            }
-        }
-
-        // Fetch from server via describe
-        let description = self.describe().await?;
-        let location = description.location.ok_or_else(|| Error::NotSupported {
-            message: "Table URI not supported by the server".into(),
-        })?;
-
-        // Cache the location for future use
-        {
-            let mut cached_location = self.location.write().await;
-            *cached_location = Some(location.clone());
-        }
-
-        Ok(location)
+    fn dataset_uri(&self) -> &str {
+        "NOT_SUPPORTED"
    }

    async fn storage_options(&self) -> Option<HashMap<String, String>> {
@@ -2232,6 +2195,7 @@ mod tests {
            let body: serde_json::Value = serde_json::from_slice(body).unwrap();
            let mut expected_body = serde_json::json!({
                "prefilter": true,
+                "distance_type": "l2",
                "nprobes": 20,
                "minimum_nprobes": 20,
                "maximum_nprobes": 20,
@@ -3357,69 +3321,4 @@ mod tests {
        let result = table.drop_columns(&["old_col1", "old_col2"]).await.unwrap();
        assert_eq!(result.version, 5);
    }
-
-    #[tokio::test]
-    async fn test_uri() {
-        let table = Table::new_with_handler("my_table", |request| {
-            assert_eq!(request.method(), "POST");
-            assert_eq!(request.url().path(), "/v1/table/my_table/describe/");
-
-            http::Response::builder()
-                .status(200)
-                .body(r#"{"version": 1, "schema": {"fields": []}, "location": "s3://bucket/path/to/table"}"#)
-                .unwrap()
-        });
-
-        let uri = table.uri().await.unwrap();
-        assert_eq!(uri, "s3://bucket/path/to/table");
-    }
-
-    #[tokio::test]
-    async fn test_uri_missing_location() {
-        let table = Table::new_with_handler("my_table", |request| {
-            assert_eq!(request.method(), "POST");
-            assert_eq!(request.url().path(), "/v1/table/my_table/describe/");
-
-            // Server returns response without location field
-            http::Response::builder()
-                .status(200)
-                .body(r#"{"version": 1, "schema": {"fields": []}}"#)
-                .unwrap()
-        });
-
-        let result = table.uri().await;
-        assert!(result.is_err());
-        assert!(matches!(&result, Err(Error::NotSupported { .. })));
-    }
-
-    #[tokio::test]
-    async fn test_uri_caching() {
-        use std::sync::atomic::{AtomicUsize, Ordering};
-        use std::sync::Arc;
-
-        let call_count = Arc::new(AtomicUsize::new(0));
-        let call_count_clone = call_count.clone();
-
-        let table = Table::new_with_handler("my_table", move |request| {
-            assert_eq!(request.url().path(), "/v1/table/my_table/describe/");
-            call_count_clone.fetch_add(1, Ordering::SeqCst);
-
-            http::Response::builder()
-                .status(200)
-                .body(
-                    r#"{"version": 1, "schema": {"fields": []}, "location": "gs://bucket/table"}"#,
-                )
-                .unwrap()
-        });
-
-        // First call should fetch from server
-        let uri1 = table.uri().await.unwrap();
-        assert_eq!(uri1, "gs://bucket/table");
-        assert_eq!(call_count.load(Ordering::SeqCst), 1);
-
-        // Second call should use cached value
-        let uri2 = table.uri().await.unwrap();
-        assert_eq!(uri2, "gs://bucket/table");
-        assert_eq!(call_count.load(Ordering::SeqCst), 1); // Still 1, no new call
-    }
 }
--- a/rust/lancedb/src/table.rs
+++ b/rust/lancedb/src/table.rs
@@ -40,10 +40,10 @@ use lance_index::vector::pq::PQBuildParams;
 use lance_index::vector::sq::builder::SQBuildParams;
 use lance_index::DatasetIndexExt;
 use lance_index::IndexType;
-use lance_io::object_store::{LanceNamespaceStorageOptionsProvider, StorageOptionsAccessor};
+use lance_io::object_store::LanceNamespaceStorageOptionsProvider;
 use lance_namespace::models::{
-    QueryTableRequest as NsQueryTableRequest, QueryTableRequestColumns,
-    QueryTableRequestFullTextQuery, QueryTableRequestVector, StringFtsQuery,
+    QueryTableRequest as NsQueryTableRequest, QueryTableRequestFullTextQuery,
+    QueryTableRequestVector, StringFtsQuery,
 };
 use lance_namespace::LanceNamespace;
 use lance_table::format::Manifest;
@@ -608,8 +608,8 @@ pub trait BaseTable: std::fmt::Display + std::fmt::Debug + Send + Sync {
    async fn list_versions(&self) -> Result<Vec<Version>>;
    /// Get the table definition.
    async fn table_definition(&self) -> Result<TableDefinition>;
-    /// Get the table URI (storage location)
-    async fn uri(&self) -> Result<String>;
+    /// Get the table URI
+    fn dataset_uri(&self) -> &str;
    /// Get the storage options used when opening this table, if any.
    async fn storage_options(&self) -> Option<HashMap<String, String>>;
    /// Poll until the columns are fully indexed. Will return Error::Timeout if the columns
@@ -1317,12 +1317,11 @@ impl Table {
        self.inner.list_indices().await
    }

-    /// Get the table URI (storage location)
+    /// Get the underlying dataset URI
    ///
-    /// Returns the full storage location of the table (e.g., S3/GCS path).
-    /// For remote tables, this fetches the location from the server via describe.
-    pub async fn uri(&self) -> Result<String> {
-        self.inner.uri().await
+    /// Warning: This is an internal API and the return value is subject to change.
+    pub fn dataset_uri(&self) -> &str {
+        self.inner.dataset_uri()
    }

    /// Get the storage options used when opening this table, if any.
@@ -1425,9 +1424,7 @@ impl Table {
            })
            .collect::<Vec<_>>();

-        let unioned = UnionExec::try_new(projected_plans).map_err(|err| Error::Runtime {
-            message: err.to_string(),
-        })?;
+        let unioned = Arc::new(UnionExec::new(projected_plans));
        // We require 1 partition in the final output
        let repartitioned = RepartitionExec::try_new(
            unioned,
@@ -1668,14 +1665,18 @@ impl NativeTable {

        // Use DatasetBuilder::from_namespace which automatically fetches location
        // and storage options from the namespace
-        let builder = DatasetBuilder::from_namespace(namespace_client.clone(), table_id)
-            .await
-            .map_err(|e| match e {
-                lance::Error::Namespace { source, .. } => Error::Runtime {
-                    message: format!("Failed to get table info from namespace: {:?}", source),
-                },
-                source => Error::Lance { source },
-            })?;
+        let builder = DatasetBuilder::from_namespace(
+            namespace_client.clone(),
+            table_id,
+            false, // Don't ignore namespace storage options
+        )
+        .await
+        .map_err(|e| match e {
+            lance::Error::Namespace { source, .. } => Error::Runtime {
+                message: format!("Failed to get table info from namespace: {:?}", source),
+            },
+            source => Error::Lance { source },
+        })?;

        let dataset = builder
            .with_read_params(params)
@@ -1879,13 +1880,7 @@ impl NativeTable {
        let store_params = params
            .store_params
            .get_or_insert_with(ObjectStoreParams::default);
-        let accessor = match store_params.storage_options().cloned() {
-            Some(options) => {
-                StorageOptionsAccessor::with_initial_and_provider(options, storage_options_provider)
-            }
-            None => StorageOptionsAccessor::with_provider(storage_options_provider),
-        };
-        store_params.storage_options_accessor = Some(Arc::new(accessor));
+        store_params.storage_options_provider = Some(storage_options_provider);

        // Patch the params if we have a write store wrapper
        let params = match write_store_wrapper.clone() {
@@ -2353,10 +2348,7 @@ impl NativeTable {
                // Convert select to columns list
                let columns = match &vq.base.select {
                    Select::All => None,
-                    Select::Columns(cols) => Some(Box::new(QueryTableRequestColumns {
-                        column_names: Some(cols.clone()),
-                        column_aliases: None,
-                    })),
+                    Select::Columns(cols) => Some(cols.clone()),
                    Select::Dynamic(_) => {
                        return Err(Error::NotSupported {
                            message:
@@ -2409,7 +2401,7 @@ impl NativeTable {
                    with_row_id: Some(vq.base.with_row_id),
                    bypass_vector_index: Some(!vq.use_index),
                    full_text_query,
-                    ..Default::default()
+                    version: None,
                })
            }
            AnyQuery::Query(q) => {
@@ -2429,10 +2421,7 @@ impl NativeTable {

                let columns = match &q.select {
                    Select::All => None,
-                    Select::Columns(cols) => Some(Box::new(QueryTableRequestColumns {
-                        column_names: Some(cols.clone()),
-                        column_aliases: None,
-                    })),
+                    Select::Columns(cols) => Some(cols.clone()),
                    Select::Dynamic(_) => {
                        return Err(Error::NotSupported {
                            message: "Dynamic columns are not supported for server-side query"
@@ -2471,11 +2460,18 @@ impl NativeTable {
                    columns,
                    prefilter: Some(q.prefilter),
                    offset: q.offset.map(|o| o as i32),
+                    ef: None,
+                    refine_factor: None,
+                    distance_type: None,
+                    nprobes: None,
                    vector_column: None, // No vector column for plain queries
                    with_row_id: Some(q.with_row_id),
                    bypass_vector_index: Some(true), // No vector index for plain queries
                    full_text_query,
-                    ..Default::default()
+                    version: None,
+                    fast_search: None,
+                    lower_bound: None,
+                    upper_bound: None,
                })
            }
        }
@@ -3228,8 +3224,8 @@ impl BaseTable for NativeTable {
        Ok(results.into_iter().flatten().collect())
    }

-    async fn uri(&self) -> Result<String> {
-        Ok(self.uri.clone())
+    fn dataset_uri(&self) -> &str {
+        self.uri.as_str()
    }

    async fn storage_options(&self) -> Option<HashMap<String, String>> {
@@ -3237,7 +3233,7 @@ impl BaseTable for NativeTable {
            .get()
            .await
            .ok()
-            .and_then(|dataset| dataset.initial_storage_options().cloned())
+            .and_then(|dataset| dataset.storage_options().cloned())
    }

    async fn index_stats(&self, index_name: &str) -> Result<Option<IndexStatistics>> {
@@ -5150,13 +5146,7 @@ mod tests {
        assert_eq!(ns_request.k, 10);
        assert_eq!(ns_request.offset, Some(5));
        assert_eq!(ns_request.filter, Some("id > 0".to_string()));
-        assert_eq!(
-            ns_request
-                .columns
-                .as_ref()
-                .and_then(|c| c.column_names.as_ref()),
-            Some(&vec!["id".to_string()])
-        );
+        assert_eq!(ns_request.columns, Some(vec!["id".to_string()]));
        assert_eq!(ns_request.vector_column, Some("vector".to_string()));
        assert_eq!(ns_request.distance_type, Some("l2".to_string()));
        assert!(ns_request.vector.single_vector.is_some());
@@ -5197,13 +5187,7 @@ mod tests {
        assert_eq!(ns_request.k, 20);
        assert_eq!(ns_request.offset, Some(5));
        assert_eq!(ns_request.filter, Some("id > 5".to_string()));
-        assert_eq!(
-            ns_request
-                .columns
-                .as_ref()
-                .and_then(|c| c.column_names.as_ref()),
-            Some(&vec!["id".to_string()])
-        );
+        assert_eq!(ns_request.columns, Some(vec!["id".to_string()]));
        assert_eq!(ns_request.with_row_id, Some(true));
        assert_eq!(ns_request.bypass_vector_index, Some(true));
        assert!(ns_request.vector_column.is_none()); // No vector column for plain queries
--- a/rust/lancedb/src/table/dataset.rs
+++ b/rust/lancedb/src/table/dataset.rs
@@ -100,8 +100,7 @@ impl DatasetRef {
                let should_checkout = match &target_ref {
                    refs::Ref::Version(_, Some(target_ver)) => version != target_ver,
                    refs::Ref::Version(_, None) => true, // No specific version, always checkout
-                    refs::Ref::VersionNumber(target_ver) => version != target_ver,
-                    refs::Ref::Tag(_) => true, // Always checkout for tags
+                    refs::Ref::Tag(_) => true,           // Always checkout for tags
                };

                if should_checkout {
--- a/rust/lancedb/src/test_utils/connection.rs
+++ b/rust/lancedb/src/test_utils/connection.rs
@@ -5,19 +5,16 @@

 use regex::Regex;
 use std::env;
-use std::process::Stdio;
-use tokio::io::{AsyncBufReadExt, BufReader};
-use tokio::process::{Child, ChildStdout, Command};
-use tokio::sync::mpsc;
+use std::io::{BufRead, BufReader};
+use std::process::{Child, ChildStdout, Command, Stdio};

 use crate::{connect, Connection};
-use anyhow::{anyhow, bail, Result};
+use anyhow::{bail, Result};
 use tempfile::{tempdir, TempDir};

 pub struct TestConnection {
    pub uri: String,
    pub connection: Connection,
-    pub is_remote: bool,
    _temp_dir: Option<TempDir>,
    _process: Option<TestProcess>,
 }
@@ -40,56 +37,6 @@ pub async fn new_test_connection() -> Result<TestConnection> {
    }
 }

-async fn spawn_stdout_reader(
-    mut stdout: BufReader<ChildStdout>,
-    port_sender: mpsc::Sender<anyhow::Result<String>>,
-) -> tokio::task::JoinHandle<()> {
-    let print_stdout = env::var("PRINT_LANCEDB_TEST_CONNECTION_SCRIPT_OUTPUT").is_ok();
-    tokio::spawn(async move {
-        let mut line = String::new();
-        let re = Regex::new(r"Query node now listening on 0.0.0.0:(.*)").unwrap();
-        loop {
-            line.clear();
-            let result = stdout.read_line(&mut line).await;
-            if let Err(err) = result {
-                port_sender
-                    .send(Err(anyhow!(
-                        "error while reading from process output: {}",
-                        err
-                    )))
-                    .await
-                    .unwrap();
-                return;
-            } else if result.unwrap() == 0 {
-                port_sender
-                    .send(Err(anyhow!(
-                        " hit EOF before reading port from process output."
-                    )))
-                    .await
-                    .unwrap();
-                return;
-            }
-            if re.is_match(&line) {
-                let caps = re.captures(&line).unwrap();
-                port_sender.send(Ok(caps[1].to_string())).await.unwrap();
-                break;
-            }
-        }
-        loop {
-            line.clear();
-            match stdout.read_line(&mut line).await {
-                Err(_) => return,
-                Ok(0) => return,
-                Ok(_size) => {
-                    if print_stdout {
-                        print!("{}", line);
-                    }
-                }
-            }
-        }
-    })
-}
-
 async fn new_remote_connection(script_path: &str) -> Result<TestConnection> {
    let temp_dir = tempdir()?;
    let data_path = temp_dir.path().to_str().unwrap().to_string();
@@ -110,25 +57,38 @@ async fn new_remote_connection(script_path: &str) -> Result<TestConnection> {
        child: child_result.unwrap(),
    };
    let stdout = BufReader::new(process.child.stdout.take().unwrap());
-    let (port_sender, mut port_receiver) = mpsc::channel(5);
-    let _reader = spawn_stdout_reader(stdout, port_sender).await;
-    let port = match port_receiver.recv().await {
-        None => bail!("Unable to determine the port number used by the phalanx process we spawned, because the reader thread was closed too soon."),
-        Some(Err(err)) => bail!("Unable to determine the port number used by the phalanx process we spawned, because of an error, {}", err),
-        Some(Ok(port)) => port,
-    };
+    let port = read_process_port(stdout)?;
    let uri = "db://test";
    let host_override = format!("http://localhost:{}", port);
    let connection = create_new_connection(uri, &host_override).await?;
    Ok(TestConnection {
        uri: uri.to_string(),
        connection,
-        is_remote: true,
        _temp_dir: Some(temp_dir),
        _process: Some(process),
    })
 }

+fn read_process_port(mut stdout: BufReader<ChildStdout>) -> Result<String> {
+    let mut line = String::new();
+    let re = Regex::new(r"Query node now listening on 0.0.0.0:(.*)").unwrap();
+    loop {
+        let result = stdout.read_line(&mut line);
+        if let Err(err) = result {
+            bail!(format!(
+                "read_process_port: error while reading from process output: {}",
+                err
+            ));
+        } else if result.unwrap() == 0 {
+            bail!("read_process_port: hit EOF before reading port from process output.");
+        }
+        if re.is_match(&line) {
+            let caps = re.captures(&line).unwrap();
+            return Ok(caps[1].to_string());
+        }
+    }
+}
+
 #[cfg(feature = "remote")]
 async fn create_new_connection(uri: &str, host_override: &str) -> crate::error::Result<Connection> {
    connect(uri)
@@ -154,7 +114,6 @@ async fn new_local_connection() -> Result<TestConnection> {
    Ok(TestConnection {
        uri: uri.to_string(),
        connection,
-        is_remote: false,
        _temp_dir: Some(temp_dir),
        _process: None,
    })
--- a/rust/lancedb/tests/embeddings_parallel_test.rs
+++ b/rust/lancedb/tests/embeddings_parallel_test.rs
@@ -1,253 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-use std::{
-    borrow::Cow,
-    sync::{
-        atomic::{AtomicUsize, Ordering},
-        Arc,
-    },
-    time::Duration,
-};
-
-use arrow::buffer::NullBuffer;
-use arrow_array::{
-    Array, FixedSizeListArray, Float32Array, RecordBatch, RecordBatchIterator, StringArray,
-};
-use arrow_schema::{DataType, Field, Schema};
-use lancedb::{
-    embeddings::{EmbeddingDefinition, EmbeddingFunction, MaybeEmbedded, WithEmbeddings},
-    Error, Result,
-};
-
-#[derive(Debug)]
-struct SlowMockEmbed {
-    name: String,
-    dim: usize,
-    delay_ms: u64,
-    call_count: Arc<AtomicUsize>,
-}
-
-impl SlowMockEmbed {
-    pub fn new(name: String, dim: usize, delay_ms: u64) -> Self {
-        Self {
-            name,
-            dim,
-            delay_ms,
-            call_count: Arc::new(AtomicUsize::new(0)),
-        }
-    }
-
-    pub fn get_call_count(&self) -> usize {
-        self.call_count.load(Ordering::SeqCst)
-    }
-}
-
-impl EmbeddingFunction for SlowMockEmbed {
-    fn name(&self) -> &str {
-        &self.name
-    }
-
-    fn source_type(&self) -> Result<Cow<'_, DataType>> {
-        Ok(Cow::Owned(DataType::Utf8))
-    }
-
-    fn dest_type(&self) -> Result<Cow<'_, DataType>> {
-        Ok(Cow::Owned(DataType::new_fixed_size_list(
-            DataType::Float32,
-            self.dim as _,
-            true,
-        )))
-    }
-
-    fn compute_source_embeddings(&self, source: Arc<dyn Array>) -> Result<Arc<dyn Array>> {
-        // Simulate slow embedding computation
-        std::thread::sleep(Duration::from_millis(self.delay_ms));
-        self.call_count.fetch_add(1, Ordering::SeqCst);
-
-        let len = source.len();
-        let inner = Arc::new(Float32Array::from(vec![Some(1.0); len * self.dim]));
-        let field = Field::new("item", inner.data_type().clone(), false);
-        let arr = FixedSizeListArray::new(
-            Arc::new(field),
-            self.dim as _,
-            inner,
-            Some(NullBuffer::new_valid(len)),
-        );
-
-        Ok(Arc::new(arr))
-    }
-
-    fn compute_query_embeddings(&self, _input: Arc<dyn Array>) -> Result<Arc<dyn Array>> {
-        unimplemented!()
-    }
-}
-
-fn create_test_batch() -> Result<RecordBatch> {
-    let schema = Arc::new(Schema::new(vec![Field::new("text", DataType::Utf8, false)]));
-    let text = StringArray::from(vec!["hello", "world"]);
-    RecordBatch::try_new(schema, vec![Arc::new(text)]).map_err(|e| Error::Runtime {
-        message: format!("Failed to create test batch: {}", e),
-    })
-}
-
-#[test]
-fn test_single_embedding_fast_path() {
-    // Single embedding should execute without spawning threads
-    let batch = create_test_batch().unwrap();
-    let schema = batch.schema();
-
-    let embed = Arc::new(SlowMockEmbed::new("test".to_string(), 2, 10));
-    let embedding_def = EmbeddingDefinition::new("text", "test", Some("embedding"));
-
-    let reader = RecordBatchIterator::new(vec![Ok(batch)], schema);
-    let embeddings = vec![(embedding_def, embed.clone() as Arc<dyn EmbeddingFunction>)];
-    let mut with_embeddings = WithEmbeddings::new(reader, embeddings);
-
-    let result = with_embeddings.next().unwrap().unwrap();
-    assert!(result.column_by_name("embedding").is_some());
-    assert_eq!(embed.get_call_count(), 1);
-}
-
-#[test]
-fn test_multiple_embeddings_parallel() {
-    // Multiple embeddings should execute in parallel
-    let batch = create_test_batch().unwrap();
-    let schema = batch.schema();
-
-    let embed1 = Arc::new(SlowMockEmbed::new("embed1".to_string(), 2, 100));
-    let embed2 = Arc::new(SlowMockEmbed::new("embed2".to_string(), 3, 100));
-    let embed3 = Arc::new(SlowMockEmbed::new("embed3".to_string(), 4, 100));
-
-    let def1 = EmbeddingDefinition::new("text", "embed1", Some("emb1"));
-    let def2 = EmbeddingDefinition::new("text", "embed2", Some("emb2"));
-    let def3 = EmbeddingDefinition::new("text", "embed3", Some("emb3"));
-
-    let reader = RecordBatchIterator::new(vec![Ok(batch)], schema);
-    let embeddings = vec![
-        (def1, embed1.clone() as Arc<dyn EmbeddingFunction>),
-        (def2, embed2.clone() as Arc<dyn EmbeddingFunction>),
-        (def3, embed3.clone() as Arc<dyn EmbeddingFunction>),
-    ];
-    let mut with_embeddings = WithEmbeddings::new(reader, embeddings);
-
-    let result = with_embeddings.next().unwrap().unwrap();
-
-    // Verify all embedding columns are present
-    assert!(result.column_by_name("emb1").is_some());
-    assert!(result.column_by_name("emb2").is_some());
-    assert!(result.column_by_name("emb3").is_some());
-
-    // Verify all embeddings were computed
-    assert_eq!(embed1.get_call_count(), 1);
-    assert_eq!(embed2.get_call_count(), 1);
-    assert_eq!(embed3.get_call_count(), 1);
-}
-
-#[test]
-fn test_embedding_column_order_preserved() {
-    // Verify that embedding columns are added in the same order as definitions
-    let batch = create_test_batch().unwrap();
-    let schema = batch.schema();
-
-    let embed1 = Arc::new(SlowMockEmbed::new("embed1".to_string(), 2, 10));
-    let embed2 = Arc::new(SlowMockEmbed::new("embed2".to_string(), 3, 10));
-    let embed3 = Arc::new(SlowMockEmbed::new("embed3".to_string(), 4, 10));
-
-    let def1 = EmbeddingDefinition::new("text", "embed1", Some("first"));
-    let def2 = EmbeddingDefinition::new("text", "embed2", Some("second"));
-    let def3 = EmbeddingDefinition::new("text", "embed3", Some("third"));
-
-    let reader = RecordBatchIterator::new(vec![Ok(batch)], schema);
-    let embeddings = vec![
-        (def1, embed1 as Arc<dyn EmbeddingFunction>),
-        (def2, embed2 as Arc<dyn EmbeddingFunction>),
-        (def3, embed3 as Arc<dyn EmbeddingFunction>),
-    ];
-    let mut with_embeddings = WithEmbeddings::new(reader, embeddings);
-
-    let result = with_embeddings.next().unwrap().unwrap();
-    let result_schema = result.schema();
-
-    // Original column is first
-    assert_eq!(result_schema.field(0).name(), "text");
-    // Embedding columns follow in order
-    assert_eq!(result_schema.field(1).name(), "first");
-    assert_eq!(result_schema.field(2).name(), "second");
-    assert_eq!(result_schema.field(3).name(), "third");
-}
-
-#[test]
-fn test_embedding_error_propagation() {
-    // Test that errors from embedding computation are properly propagated
-    #[derive(Debug)]
-    struct FailingEmbed {
-        name: String,
-    }
-
-    impl EmbeddingFunction for FailingEmbed {
-        fn name(&self) -> &str {
-            &self.name
-        }
-
-        fn source_type(&self) -> Result<Cow<'_, DataType>> {
-            Ok(Cow::Owned(DataType::Utf8))
-        }
-
-        fn dest_type(&self) -> Result<Cow<'_, DataType>> {
-            Ok(Cow::Owned(DataType::new_fixed_size_list(
-                DataType::Float32,
-                2,
-                true,
-            )))
-        }
-
-        fn compute_source_embeddings(&self, _source: Arc<dyn Array>) -> Result<Arc<dyn Array>> {
-            Err(Error::Runtime {
-                message: "Intentional failure".to_string(),
-            })
-        }
-
-        fn compute_query_embeddings(&self, _input: Arc<dyn Array>) -> Result<Arc<dyn Array>> {
-            unimplemented!()
-        }
-    }
-
-    let batch = create_test_batch().unwrap();
-    let schema = batch.schema();
-
-    let embed = Arc::new(FailingEmbed {
-        name: "failing".to_string(),
-    });
-    let def = EmbeddingDefinition::new("text", "failing", Some("emb"));
-
-    let reader = RecordBatchIterator::new(vec![Ok(batch)], schema);
-    let embeddings = vec![(def, embed as Arc<dyn EmbeddingFunction>)];
-    let mut with_embeddings = WithEmbeddings::new(reader, embeddings);
-
-    let result = with_embeddings.next().unwrap();
-    assert!(result.is_err());
-    let err_msg = format!("{}", result.err().unwrap());
-    assert!(err_msg.contains("Intentional failure"));
-}
-
-#[test]
-fn test_maybe_embedded_with_no_embeddings() {
-    // Test that MaybeEmbedded::No variant works correctly
-    let batch = create_test_batch().unwrap();
-    let schema = batch.schema();
-
-    let reader = RecordBatchIterator::new(vec![Ok(batch.clone())], schema.clone());
-    let table_def = lancedb::table::TableDefinition {
-        schema: schema.clone(),
-        column_definitions: vec![lancedb::table::ColumnDefinition {
-            kind: lancedb::table::ColumnKind::Physical,
-        }],
-    };
-
-    let mut maybe_embedded = MaybeEmbedded::try_new(reader, table_def, None).unwrap();
-
-    let result = maybe_embedded.next().unwrap().unwrap();
-    assert_eq!(result.num_columns(), 1);
-    assert_eq!(result.column(0).as_ref(), batch.column(0).as_ref());
-}