Merge branch 'main' into openai-vector-store/qdrant

2025-08-15 06:00:48 +00:00 · 2025-08-01 00:39:03 -04:00 · 2025-08-01 00:39:03 -04:00 · 354f79f15c
commit 354f79f15c
parent 970d0f307f 194abe7734
20 changed files with 2411 additions and 2137 deletions
--- a/.github/workflows/README.md
+++ b/.github/workflows/README.md
@ -13,6 +13,7 @@ Llama Stack uses GitHub Actions for Continuous Integration (CI). Below is a tabl
 | Pre-commit | [pre-commit.yml](pre-commit.yml) | Run pre-commit checks |
 | Test Llama Stack Build | [providers-build.yml](providers-build.yml) | Test llama stack build |
 | Python Package Build Test | [python-build-test.yml](python-build-test.yml) | Test building the llama-stack PyPI project |
+| Integration Tests (Record) | [record-integration-tests.yml](record-integration-tests.yml) | Run the integration test suite from tests/integration |
 | Check semantic PR titles | [semantic-pr.yml](semantic-pr.yml) | Ensure that PR titles follow the conventional commit spec |
 | Close stale issues and PRs | [stale_bot.yml](stale_bot.yml) | Run the Stale Bot action |
 | Test External Providers Installed via Module | [test-external-provider-module.yml](test-external-provider-module.yml) | Test External Provider installation via Python module |
--- a/.github/workflows/integration-vector-io-tests.yml
+++ b/.github/workflows/integration-vector-io-tests.yml
@ -24,7 +24,7 @@ jobs:
    runs-on: ubuntu-latest
    strategy:
      matrix:
-        vector-io-provider: ["inline::faiss", "inline::sqlite-vec", "inline::milvus", "remote::chromadb", "remote::pgvector", "remote::qdrant"]
+        vector-io-provider: ["inline::faiss", "inline::sqlite-vec", "inline::milvus", "remote::chromadb", "remote::pgvector", "remote::weaviate", "remote::qdrant"]
        python-version: ["3.12", "3.13"]
      fail-fast: false # we want to run all tests regardless of failure

@ -48,6 +48,14 @@ jobs:
            -e ANONYMIZED_TELEMETRY=FALSE \
            chromadb/chroma:latest

+      - name: Setup Weaviate
+        if: matrix.vector-io-provider == 'remote::weaviate'
+        run: |
+          docker run --rm -d --pull always \
+          --name weaviate \
+          -p 8080:8080 -p 50051:50051 \
+          cr.weaviate.io/semitechnologies/weaviate:1.32.0
+
      - name: Start PGVector DB
        if: matrix.vector-io-provider == 'remote::pgvector'
        run: |
@ -116,6 +124,21 @@ jobs:
          docker logs chromadb
          exit 1

+      - name: Wait for Weaviate to be ready
+        if: matrix.vector-io-provider == 'remote::weaviate'
+        run: |
+          echo "Waiting for Weaviate to be ready..."
+          for i in {1..30}; do
+            if curl -s http://localhost:8080 | grep -q "https://weaviate.io/developers/weaviate/current/"; then
+              echo "Weaviate is ready!"
+              exit 0
+            fi
+            sleep 2
+          done
+          echo "Weaviate failed to start"
+          docker logs weaviate
+          exit 1
+
      - name: Build Llama Stack
        run: |
          uv run llama stack build --template ci-tests --image-type venv
@ -138,6 +161,8 @@ jobs:
          PGVECTOR_PASSWORD: ${{ matrix.vector-io-provider == 'remote::pgvector' && 'llamastack' || '' }}
          ENABLE_QDRANT: ${{ matrix.vector-io-provider == 'remote::qdrant' && 'true' || '' }}
          QDRANT_URL: ${{ matrix.vector-io-provider == 'remote::qdrant' && 'http://localhost:6333' || '' }}
+          ENABLE_WEAVIATE: ${{ matrix.vector-io-provider == 'remote::weaviate' && 'true' || '' }}
+          WEAVIATE_CLUSTER_URL: ${{ matrix.vector-io-provider == 'remote::weaviate' && 'localhost:8080' || '' }}
        run: |
          uv run pytest -sv --stack-config="inference=inline::sentence-transformers,vector_io=${{ matrix.vector-io-provider }}" \
            tests/integration/vector_io \
--- a/.github/workflows/record-integration-tests.yml
+++ b/.github/workflows/record-integration-tests.yml
@ -0,0 +1,109 @@
+name: Integration Tests (Record)
+
+run-name: Run the integration test suite from tests/integration
+
+on:
+  pull_request:
+    branches: [ main ]
+    types: [opened, synchronize, labeled]
+    paths:
+      - 'llama_stack/**'
+      - 'tests/**'
+      - 'uv.lock'
+      - 'pyproject.toml'
+      - '.github/workflows/record-integration-tests.yml' # This workflow
+      - '.github/actions/setup-ollama/action.yml'
+      - '.github/actions/setup-test-environment/action.yml'
+      - '.github/actions/run-and-record-tests/action.yml'
+  workflow_dispatch:
+    inputs:
+      test-provider:
+        description: 'Test against a specific provider'
+        type: string
+        default: 'ollama'
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  discover-tests:
+    if: contains(github.event.pull_request.labels.*.name, 're-record-tests') ||
+      contains(github.event.pull_request.labels.*.name, 're-record-vision-tests')
+    runs-on: ubuntu-latest
+    outputs:
+      test-types: ${{ steps.generate-test-types.outputs.test-types }}
+      matrix-modes: ${{ steps.generate-test-types.outputs.matrix-modes }}
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+
+      - name: Generate test types
+        id: generate-test-types
+        run: |
+          # Get test directories dynamically, excluding non-test directories
+          TEST_TYPES=$(find tests/integration -maxdepth 1 -mindepth 1 -type d -printf "%f\n" |
+            grep -Ev "^(__pycache__|fixtures|test_cases|recordings|post_training)$" |
+            sort | jq -R -s -c 'split("\n")[:-1]')
+          echo "test-types=$TEST_TYPES" >> $GITHUB_OUTPUT
+
+          labels=$(gh pr view ${{ github.event.pull_request.number }} --json labels --jq '.labels[].name')
+          echo "labels=$labels"
+
+          modes_array=()
+          if [[ $labels == *"re-record-vision-tests"* ]]; then
+            modes_array+=("vision")
+          fi
+          if [[ $labels == *"re-record-tests"* ]]; then
+            modes_array+=("non-vision")
+          fi
+
+          # Convert to JSON array
+          if [ ${#modes_array[@]} -eq 0 ]; then
+            matrix_modes="[]"
+          else
+            matrix_modes=$(printf '%s\n' "${modes_array[@]}" | jq -R -s -c 'split("\n")[:-1]')
+          fi
+          echo "matrix_modes=$matrix_modes"
+          echo "matrix-modes=$matrix_modes" >> $GITHUB_OUTPUT
+
+        env:
+          GH_TOKEN: ${{ github.token }}
+
+  record-tests:
+    needs: discover-tests
+    runs-on: ubuntu-latest
+
+    permissions:
+      contents: write
+
+    strategy:
+      fail-fast: false
+      matrix:
+        mode: ${{ fromJSON(needs.discover-tests.outputs.matrix-modes) }}
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+        with:
+          ref: ${{ github.event.pull_request.head.ref }}
+          fetch-depth: 0
+
+      - name: Setup test environment
+        uses: ./.github/actions/setup-test-environment
+        with:
+          python-version: "3.12"  # Use single Python version for recording
+          client-version: "latest"
+          provider: ${{ inputs.test-provider || 'ollama' }}
+          run-vision-tests: ${{ matrix.mode == 'vision' && 'true' || 'false' }}
+          inference-mode: 'record'
+
+      - name: Run and record tests
+        uses: ./.github/actions/run-and-record-tests
+        with:
+          test-types: ${{ needs.discover-tests.outputs.test-types }}
+          stack-config: 'server:ci-tests'  # recording must be done with server since more tests are run
+          provider: ${{ inputs.test-provider || 'ollama' }}
+          inference-mode: 'record'
+          run-vision-tests: ${{ matrix.mode == 'vision' && 'true' || 'false' }}
--- a/docs/source/providers/vector_io/remote_weaviate.md
+++ b/docs/source/providers/vector_io/remote_weaviate.md
@ -33,9 +33,19 @@ To install Weaviate see the [Weaviate quickstart documentation](https://weaviate
 See [Weaviate's documentation](https://weaviate.io/developers/weaviate) for more details about Weaviate in general.


+## Configuration
+
+| Field | Type | Required | Default | Description |
+|-------|------|----------|---------|-------------|
+| `weaviate_api_key` | `str \| None` | No |  | The API key for the Weaviate instance |
+| `weaviate_cluster_url` | `str \| None` | No | localhost:8080 | The URL of the Weaviate cluster |
+| `kvstore` | `utils.kvstore.config.RedisKVStoreConfig \| utils.kvstore.config.SqliteKVStoreConfig \| utils.kvstore.config.PostgresKVStoreConfig \| utils.kvstore.config.MongoDBKVStoreConfig, annotation=NoneType, required=False, default='sqlite', discriminator='type'` | No |  | Config for KV store backend (SQLite only for now) |
+
 ## Sample Configuration

 ```yaml
+weaviate_api_key: null
+weaviate_cluster_url: ${env.WEAVIATE_CLUSTER_URL:=localhost:8080}
 kvstore:
  type: sqlite
  db_path: ${env.SQLITE_STORE_DIR:=~/.llama/dummy}/weaviate_registry.db
--- a/llama_stack/apis/vector_io/vector_io.py
+++ b/llama_stack/apis/vector_io/vector_io.py
@ -16,7 +16,7 @@ from pydantic import BaseModel, Field
 from llama_stack.apis.inference import InterleavedContent
 from llama_stack.apis.vector_dbs import VectorDB
 from llama_stack.providers.utils.telemetry.trace_protocol import trace_protocol
-from llama_stack.providers.utils.vector_io.chunk_utils import generate_chunk_id
+from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id
 from llama_stack.schema_utils import json_schema_type, webmethod
 from llama_stack.strong_typing.schema import register_schema

--- a/llama_stack/providers/remote/vector_io/milvus/milvus.py
+++ b/llama_stack/providers/remote/vector_io/milvus/milvus.py
@ -7,7 +7,6 @@
 import asyncio
 import logging
 import os
-import re
 from typing import Any

 from numpy.typing import NDArray
@ -31,6 +30,7 @@ from llama_stack.providers.utils.memory.vector_store import (
    EmbeddingIndex,
    VectorDBWithIndex,
 )
+from llama_stack.providers.utils.vector_io.vector_utils import sanitize_collection_name

 from .config import MilvusVectorIOConfig as RemoteMilvusVectorIOConfig

@ -44,14 +44,6 @@ OPENAI_VECTOR_STORES_FILES_PREFIX = f"openai_vector_stores_files:milvus:{VERSION
 OPENAI_VECTOR_STORES_FILES_CONTENTS_PREFIX = f"openai_vector_stores_files_contents:milvus:{VERSION}::"


-def sanitize_collection_name(name: str) -> str:
-    """
-    Sanitize collection name to ensure it only contains numbers, letters, and underscores.
-    Any other characters are replaced with underscores.
-    """
-    return re.sub(r"[^a-zA-Z0-9_]", "_", name)
-
-
 class MilvusIndex(EmbeddingIndex):
    def __init__(
        self, client: MilvusClient, collection_name: str, consistency_level="Strong", kvstore: KVStore | None = None
--- a/llama_stack/providers/remote/vector_io/weaviate/init.py
+++ b/llama_stack/providers/remote/vector_io/weaviate/init.py
@ -12,6 +12,6 @@ from .config import WeaviateVectorIOConfig
 async def get_adapter_impl(config: WeaviateVectorIOConfig, deps: dict[Api, ProviderSpec]):
    from .weaviate import WeaviateVectorIOAdapter

-    impl = WeaviateVectorIOAdapter(config, deps[Api.inference])
+    impl = WeaviateVectorIOAdapter(config, deps[Api.inference], deps.get(Api.files, None))
    await impl.initialize()
    return impl
--- a/llama_stack/providers/remote/vector_io/weaviate/config.py
+++ b/llama_stack/providers/remote/vector_io/weaviate/config.py
@ -12,18 +12,24 @@ from llama_stack.providers.utils.kvstore.config import (
    KVStoreConfig,
    SqliteKVStoreConfig,
 )
+from llama_stack.schema_utils import json_schema_type


-class WeaviateRequestProviderData(BaseModel):
-    weaviate_api_key: str
-    weaviate_cluster_url: str
+@json_schema_type
+class WeaviateVectorIOConfig(BaseModel):
+    weaviate_api_key: str | None = Field(description="The API key for the Weaviate instance", default=None)
+    weaviate_cluster_url: str | None = Field(description="The URL of the Weaviate cluster", default="localhost:8080")
    kvstore: KVStoreConfig | None = Field(description="Config for KV store backend (SQLite only for now)", default=None)

-
-class WeaviateVectorIOConfig(BaseModel):
    @classmethod
-    def sample_run_config(cls, __distro_dir__: str, **kwargs: Any) -> dict[str, Any]:
+    def sample_run_config(
+        cls,
+        __distro_dir__: str,
+        **kwargs: Any,
+    ) -> dict[str, Any]:
        return {
+            "weaviate_api_key": None,
+            "weaviate_cluster_url": "${env.WEAVIATE_CLUSTER_URL:=localhost:8080}",
            "kvstore": SqliteKVStoreConfig.sample_run_config(
                __distro_dir__=__distro_dir__,
                db_name="weaviate_registry.db",
--- a/llama_stack/providers/remote/vector_io/weaviate/weaviate.py
+++ b/llama_stack/providers/remote/vector_io/weaviate/weaviate.py
@ -22,12 +22,16 @@ from llama_stack.core.request_headers import NeedsRequestProviderData
 from llama_stack.providers.datatypes import Api, VectorDBsProtocolPrivate
 from llama_stack.providers.utils.kvstore import kvstore_impl
 from llama_stack.providers.utils.kvstore.api import KVStore
+from llama_stack.providers.utils.memory.openai_vector_store_mixin import (
+    OpenAIVectorStoreMixin,
+)
 from llama_stack.providers.utils.memory.vector_store import (
    EmbeddingIndex,
    VectorDBWithIndex,
 )
+from llama_stack.providers.utils.vector_io.vector_utils import sanitize_collection_name

-from .config import WeaviateRequestProviderData, WeaviateVectorIOConfig
+from .config import WeaviateVectorIOConfig

 log = logging.getLogger(__name__)

@ -40,11 +44,19 @@ OPENAI_VECTOR_STORES_FILES_CONTENTS_PREFIX = f"openai_vector_stores_files_conten


 class WeaviateIndex(EmbeddingIndex):
-    def __init__(self, client: weaviate.Client, collection_name: str, kvstore: KVStore | None = None):
+    def __init__(
+        self,
+        client: weaviate.Client,
+        collection_name: str,
+        kvstore: KVStore | None = None,
+    ):
        self.client = client
-        self.collection_name = collection_name
+        self.collection_name = sanitize_collection_name(collection_name, weaviate_format=True)
        self.kvstore = kvstore

+    async def initialize(self):
+        pass
+
    async def add_chunks(self, chunks: list[Chunk], embeddings: NDArray):
        assert len(chunks) == len(embeddings), (
            f"Chunk length {len(chunks)} does not match embedding length {len(embeddings)}"
@ -68,10 +80,13 @@ class WeaviateIndex(EmbeddingIndex):
        collection.data.insert_many(data_objects)

    async def delete_chunk(self, chunk_id: str) -> None:
-        raise NotImplementedError("delete_chunk is not supported in Chroma")
+        sanitized_collection_name = sanitize_collection_name(self.collection_name, weaviate_format=True)
+        collection = self.client.collections.get(sanitized_collection_name)
+        collection.data.delete_many(where=Filter.by_property("id").contains_any([chunk_id]))

    async def query_vector(self, embedding: NDArray, k: int, score_threshold: float) -> QueryChunksResponse:
-        collection = self.client.collections.get(self.collection_name)
+        sanitized_collection_name = sanitize_collection_name(self.collection_name, weaviate_format=True)
+        collection = self.client.collections.get(sanitized_collection_name)

        results = collection.query.near_vector(
            near_vector=embedding.tolist(),
@ -95,8 +110,17 @@ class WeaviateIndex(EmbeddingIndex):

        return QueryChunksResponse(chunks=chunks, scores=scores)

-    async def delete(self, chunk_ids: list[str]) -> None:
-        collection = self.client.collections.get(self.collection_name)
+    async def delete(self, chunk_ids: list[str] | None = None) -> None:
+        """
+        Delete chunks by IDs if provided, otherwise drop the entire collection.
+        """
+        sanitized_collection_name = sanitize_collection_name(self.collection_name, weaviate_format=True)
+        if chunk_ids is None:
+            # Drop entire collection if it exists
+            if self.client.collections.exists(sanitized_collection_name):
+                self.client.collections.delete(sanitized_collection_name)
+            return
+        collection = self.client.collections.get(sanitized_collection_name)
        collection.data.delete_many(where=Filter.by_property("id").contains_any(chunk_ids))

    async def query_keyword(
@ -120,6 +144,7 @@ class WeaviateIndex(EmbeddingIndex):


 class WeaviateVectorIOAdapter(
+    OpenAIVectorStoreMixin,
    VectorIO,
    NeedsRequestProviderData,
    VectorDBsProtocolPrivate,
@ -141,34 +166,48 @@ class WeaviateVectorIOAdapter(
        self.metadata_collection_name = "openai_vector_stores_metadata"

    def _get_client(self) -> weaviate.Client:
-        provider_data = self.get_request_provider_data()
-        assert provider_data is not None, "Request provider data must be set"
-        assert isinstance(provider_data, WeaviateRequestProviderData)
-
-        key = f"{provider_data.weaviate_cluster_url}::{provider_data.weaviate_api_key}"
+        if "localhost" in self.config.weaviate_cluster_url:
+            log.info("using Weaviate locally in container")
+            host, port = self.config.weaviate_cluster_url.split(":")
+            key = "local_test"
+            client = weaviate.connect_to_local(
+                host=host,
+                port=port,
+            )
+        else:
+            log.info("Using Weaviate remote cluster with URL")
+            key = f"{self.config.weaviate_cluster_url}::{self.config.weaviate_api_key}"
            if key in self.client_cache:
                return self.client_cache[key]
-
            client = weaviate.connect_to_weaviate_cloud(
-            cluster_url=provider_data.weaviate_cluster_url,
-            auth_credentials=Auth.api_key(provider_data.weaviate_api_key),
+                cluster_url=self.config.weaviate_cluster_url,
+                auth_credentials=Auth.api_key(self.config.weaviate_api_key),
            )
        self.client_cache[key] = client
        return client

    async def initialize(self) -> None:
        """Set up KV store and load existing vector DBs and OpenAI vector stores."""
-        # Initialize KV store for metadata
+        # Initialize KV store for metadata if configured
+        if self.config.kvstore is not None:
            self.kvstore = await kvstore_impl(self.config.kvstore)
+        else:
+            self.kvstore = None
+            log.info("No kvstore configured, registry will not persist across restarts")

        # Load existing vector DB definitions
+        if self.kvstore is not None:
            start_key = VECTOR_DBS_PREFIX
            end_key = f"{VECTOR_DBS_PREFIX}\xff"
            stored = await self.kvstore.values_in_range(start_key, end_key)
            for raw in stored:
                vector_db = VectorDB.model_validate_json(raw)
                client = self._get_client()
-            idx = WeaviateIndex(client=client, collection_name=vector_db.identifier, kvstore=self.kvstore)
+                idx = WeaviateIndex(
+                    client=client,
+                    collection_name=vector_db.identifier,
+                    kvstore=self.kvstore,
+                )
                self.cache[vector_db.identifier] = VectorDBWithIndex(
                    vector_db=vector_db,
                    index=idx,
@ -187,11 +226,11 @@ class WeaviateVectorIOAdapter(
        vector_db: VectorDB,
    ) -> None:
        client = self._get_client()
-
+        sanitized_collection_name = sanitize_collection_name(vector_db.identifier, weaviate_format=True)
        # Create collection if it doesn't exist
-        if not client.collections.exists(vector_db.identifier):
+        if not client.collections.exists(sanitized_collection_name):
            client.collections.create(
-                name=vector_db.identifier,
+                name=sanitized_collection_name,
                vectorizer_config=wvc.config.Configure.Vectorizer.none(),
                properties=[
                    wvc.config.Property(
@ -201,30 +240,41 @@ class WeaviateVectorIOAdapter(
                ],
            )

-        self.cache[vector_db.identifier] = VectorDBWithIndex(
+        self.cache[sanitized_collection_name] = VectorDBWithIndex(
            vector_db,
-            WeaviateIndex(client=client, collection_name=vector_db.identifier),
+            WeaviateIndex(client=client, collection_name=sanitized_collection_name),
            self.inference_api,
        )

-    async def _get_and_cache_vector_db_index(self, vector_db_id: str) -> VectorDBWithIndex | None:
-        if vector_db_id in self.cache:
-            return self.cache[vector_db_id]
+    async def unregister_vector_db(self, vector_db_id: str) -> None:
+        client = self._get_client()
+        sanitized_collection_name = sanitize_collection_name(vector_db_id, weaviate_format=True)
+        if sanitized_collection_name not in self.cache or client.collections.exists(sanitized_collection_name) is False:
+            log.warning(f"Vector DB {sanitized_collection_name} not found")
+            return
+        client.collections.delete(sanitized_collection_name)
+        await self.cache[sanitized_collection_name].index.delete()
+        del self.cache[sanitized_collection_name]

-        vector_db = await self.vector_db_store.get_vector_db(vector_db_id)
+    async def _get_and_cache_vector_db_index(self, vector_db_id: str) -> VectorDBWithIndex | None:
+        sanitized_collection_name = sanitize_collection_name(vector_db_id, weaviate_format=True)
+        if sanitized_collection_name in self.cache:
+            return self.cache[sanitized_collection_name]
+
+        vector_db = await self.vector_db_store.get_vector_db(sanitized_collection_name)
        if not vector_db:
            raise VectorStoreNotFoundError(vector_db_id)

        client = self._get_client()
        if not client.collections.exists(vector_db.identifier):
-            raise ValueError(f"Collection with name `{vector_db.identifier}` not found")
+            raise ValueError(f"Collection with name `{sanitized_collection_name}` not found")

        index = VectorDBWithIndex(
            vector_db=vector_db,
-            index=WeaviateIndex(client=client, collection_name=vector_db.identifier),
+            index=WeaviateIndex(client=client, collection_name=sanitized_collection_name),
            inference_api=self.inference_api,
        )
-        self.cache[vector_db_id] = index
+        self.cache[sanitized_collection_name] = index
        return index

    async def insert_chunks(
@ -233,7 +283,8 @@ class WeaviateVectorIOAdapter(
        chunks: list[Chunk],
        ttl_seconds: int | None = None,
    ) -> None:
-        index = await self._get_and_cache_vector_db_index(vector_db_id)
+        sanitized_collection_name = sanitize_collection_name(vector_db_id, weaviate_format=True)
+        index = await self._get_and_cache_vector_db_index(sanitized_collection_name)
        if not index:
            raise VectorStoreNotFoundError(vector_db_id)

@ -245,29 +296,17 @@ class WeaviateVectorIOAdapter(
        query: InterleavedContent,
        params: dict[str, Any] | None = None,
    ) -> QueryChunksResponse:
-        index = await self._get_and_cache_vector_db_index(vector_db_id)
+        sanitized_collection_name = sanitize_collection_name(vector_db_id, weaviate_format=True)
+        index = await self._get_and_cache_vector_db_index(sanitized_collection_name)
        if not index:
            raise VectorStoreNotFoundError(vector_db_id)

        return await index.query_chunks(query, params)

-    # OpenAI Vector Stores File operations are not supported in Weaviate
-    async def _save_openai_vector_store_file(
-        self, store_id: str, file_id: str, file_info: dict[str, Any], file_contents: list[dict[str, Any]]
-    ) -> None:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
-
-    async def _load_openai_vector_store_file(self, store_id: str, file_id: str) -> dict[str, Any]:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
-
-    async def _load_openai_vector_store_file_contents(self, store_id: str, file_id: str) -> list[dict[str, Any]]:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
-
-    async def _update_openai_vector_store_file(self, store_id: str, file_id: str, file_info: dict[str, Any]) -> None:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
-
-    async def _delete_openai_vector_store_file_from_storage(self, store_id: str, file_id: str) -> None:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
-
    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Weaviate")
+        sanitized_collection_name = sanitize_collection_name(store_id, weaviate_format=True)
+        index = await self._get_and_cache_vector_db_index(sanitized_collection_name)
+        if not index:
+            raise ValueError(f"Vector DB {sanitized_collection_name} not found")
+
+        await index.delete(chunk_ids)
--- a/llama_stack/providers/utils/memory/vector_store.py
+++ b/llama_stack/providers/utils/memory/vector_store.py
@ -30,7 +30,7 @@ from llama_stack.providers.datatypes import Api
 from llama_stack.providers.utils.inference.prompt_adapter import (
    interleaved_content_as_str,
 )
-from llama_stack.providers.utils.vector_io.chunk_utils import generate_chunk_id
+from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id

 log = logging.getLogger(__name__)

--- a/llama_stack/providers/utils/vector_io/vector_utils.py
+++ b/llama_stack/providers/utils/vector_io/vector_utils.py
@ -5,6 +5,7 @@
 # the root directory of this source tree.

 import hashlib
+import re
 import uuid


@ -19,3 +20,20 @@ def generate_chunk_id(document_id: str, chunk_text: str, chunk_window: str | Non
    if chunk_window:
        hash_input += f":{chunk_window}".encode()
    return str(uuid.UUID(hashlib.md5(hash_input, usedforsecurity=False).hexdigest()))
+
+
+def proper_case(s: str) -> str:
+    """Convert a string to proper case (first letter uppercase, rest lowercase)."""
+    return s[0].upper() + s[1:].lower() if s else s
+
+
+def sanitize_collection_name(name: str, weaviate_format=False) -> str:
+    """
+    Sanitize collection name to ensure it only contains numbers, letters, and underscores.
+    Any other characters are replaced with underscores.
+    """
+    if not weaviate_format:
+        s = re.sub(r"[^a-zA-Z0-9_]", "_", name)
+    else:
+        s = proper_case(re.sub(r"[^a-zA-Z0-9]", "", name))
+    return s
--- a/pyproject.toml
+++ b/pyproject.toml
@ -116,6 +116,7 @@ test = [
    "requests",
    "pymilvus>=2.5.12",
    "reportlab",
+    "weaviate-client>=4.16.4",
 ]
 docs = [
    "setuptools",
--- a/tests/integration/fixtures/common.py
+++ b/tests/integration/fixtures/common.py
@ -6,6 +6,7 @@

 import inspect
 import os
+import shlex
 import signal
 import socket
 import subprocess
@ -38,10 +39,10 @@ def is_port_available(port: int, host: str = "localhost") -> bool:

 def start_llama_stack_server(config_name: str) -> subprocess.Popen:
    """Start a llama stack server with the given config."""
-    cmd = ["llama", "stack", "run", config_name]
+    cmd = f"uv run --with llama-stack llama stack build --template {config_name} --image-type venv --run"
    devnull = open(os.devnull, "w")
    process = subprocess.Popen(
-        cmd,
+        shlex.split(cmd),
        stdout=devnull,  # redirect stdout to devnull to prevent deadlock
        stderr=subprocess.PIPE,  # keep stderr to see errors
        text=True,
--- a/tests/integration/recordings/vision/index.sqlite
+++ b/tests/integration/recordings/vision/index.sqlite
--- a/tests/integration/recordings/vision/responses/4a3a4447b16b.json
+++ b/tests/integration/recordings/vision/responses/4a3a4447b16b.json
@ -14,7 +14,7 @@
        "models": [
          {
            "model": "nomic-embed-text:latest",
-            "modified_at": "2025-07-31T17:55:10.302292Z",
+            "modified_at": "2025-07-31T23:55:40.635067Z",
            "digest": "0a109f422b47e3a30ba2b10eca18548e944e8a23073ee3f3e947efcf3c45e59f",
            "size": 274302450,
            "details": {
--- a/tests/integration/recordings/vision/responses/f1592dee71e5.json
+++ b/tests/integration/recordings/vision/responses/f1592dee71e5.json
@ -30,15 +30,15 @@
      "__type__": "ollama._types.ChatResponse",
      "__data__": {
        "model": "llama3.2-vision:11b",
-        "created_at": "2025-07-31T18:05:27.306789902Z",
+        "created_at": "2025-08-01T00:06:12.068973125Z",
        "done": true,
        "done_reason": "stop",
-        "total_duration": 40783180279,
-        "load_duration": 50835227,
+        "total_duration": 44793549354,
+        "load_duration": 51960915,
        "prompt_eval_count": 18,
-        "prompt_eval_duration": 518613186,
+        "prompt_eval_duration": 579363429,
        "eval_count": 110,
-        "eval_duration": 40206118132,
+        "eval_duration": 44156162976,
        "message": {
          "role": "assistant",
          "content": "The image features a close-up of a golden retriever puppy, with its mouth open and tongue out, as if it is smiling or panting. The puppy's fur is a light golden color, and its ears are floppy and hanging down on either side of its head. The background of the image is blurred, but it appears to be a natural setting, possibly a field or a park, with a greenish-yellow color. The overall atmosphere of the image is one of happiness and playfulness, as the puppy seems to be enjoying itself.",
--- a/tests/integration/recordings/vision/responses/ff7db0102b28.json
+++ b/tests/integration/recordings/vision/responses/ff7db0102b28.json
@ -31,7 +31,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:11.793039554Z",
+          "created_at": "2025-08-01T00:04:49.339347876Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -53,7 +53,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:12.165542521Z",
+          "created_at": "2025-08-01T00:04:49.747466769Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -75,7 +75,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:12.543284243Z",
+          "created_at": "2025-08-01T00:04:50.156146804Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -97,7 +97,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:12.916189133Z",
+          "created_at": "2025-08-01T00:04:50.566195243Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -119,7 +119,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:13.28189993Z",
+          "created_at": "2025-08-01T00:04:50.975121211Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -141,7 +141,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:13.653725224Z",
+          "created_at": "2025-08-01T00:04:51.388779549Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -163,7 +163,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:14.018988652Z",
+          "created_at": "2025-08-01T00:04:51.79897453Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -185,7 +185,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:14.382645858Z",
+          "created_at": "2025-08-01T00:04:52.209608504Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -207,7 +207,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:14.739580312Z",
+          "created_at": "2025-08-01T00:04:52.619045995Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -229,7 +229,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:15.118228755Z",
+          "created_at": "2025-08-01T00:04:53.026501007Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -251,7 +251,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:15.499394357Z",
+          "created_at": "2025-08-01T00:04:53.436015071Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -273,7 +273,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:15.862659022Z",
+          "created_at": "2025-08-01T00:04:53.843369446Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -295,7 +295,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:16.234769542Z",
+          "created_at": "2025-08-01T00:04:54.255794451Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -317,7 +317,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:16.607886498Z",
+          "created_at": "2025-08-01T00:04:54.663263793Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -339,7 +339,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:16.971019788Z",
+          "created_at": "2025-08-01T00:04:55.073162133Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -361,7 +361,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:17.341647677Z",
+          "created_at": "2025-08-01T00:04:55.48667439Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -383,7 +383,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:17.7063123Z",
+          "created_at": "2025-08-01T00:04:55.897947147Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -405,7 +405,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:18.072790228Z",
+          "created_at": "2025-08-01T00:04:56.31639321Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -427,7 +427,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:18.439855563Z",
+          "created_at": "2025-08-01T00:04:56.729288843Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -449,7 +449,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:18.802891341Z",
+          "created_at": "2025-08-01T00:04:57.142647132Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -471,7 +471,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:19.167811769Z",
+          "created_at": "2025-08-01T00:04:57.55091814Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -493,7 +493,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:19.538569316Z",
+          "created_at": "2025-08-01T00:04:57.959494633Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -515,7 +515,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:19.900014289Z",
+          "created_at": "2025-08-01T00:04:58.367117419Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -537,7 +537,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:20.296603174Z",
+          "created_at": "2025-08-01T00:04:58.77560425Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -559,7 +559,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:20.670619107Z",
+          "created_at": "2025-08-01T00:04:59.183890868Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -581,7 +581,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:21.037515143Z",
+          "created_at": "2025-08-01T00:04:59.596163097Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -603,7 +603,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:21.41078269Z",
+          "created_at": "2025-08-01T00:05:00.004002773Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -625,7 +625,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:21.773638843Z",
+          "created_at": "2025-08-01T00:05:00.410717383Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -647,7 +647,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:22.134453183Z",
+          "created_at": "2025-08-01T00:05:00.817783323Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -669,7 +669,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:22.494472509Z",
+          "created_at": "2025-08-01T00:05:01.223523865Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -691,7 +691,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:22.863772543Z",
+          "created_at": "2025-08-01T00:05:01.63351174Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -713,7 +713,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:23.231941325Z",
+          "created_at": "2025-08-01T00:05:02.032702205Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -735,7 +735,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:23.592854692Z",
+          "created_at": "2025-08-01T00:05:02.424431407Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -757,7 +757,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:23.959965309Z",
+          "created_at": "2025-08-01T00:05:02.81524835Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -779,7 +779,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:24.322278389Z",
+          "created_at": "2025-08-01T00:05:03.207597567Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -801,7 +801,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:24.690079541Z",
+          "created_at": "2025-08-01T00:05:03.614094549Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -823,7 +823,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:25.080109961Z",
+          "created_at": "2025-08-01T00:05:04.008232462Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -845,7 +845,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:25.501052051Z",
+          "created_at": "2025-08-01T00:05:04.411085956Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -867,7 +867,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:25.874257383Z",
+          "created_at": "2025-08-01T00:05:04.80616608Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -889,7 +889,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:26.240808197Z",
+          "created_at": "2025-08-01T00:05:05.212911563Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -911,7 +911,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:26.633293446Z",
+          "created_at": "2025-08-01T00:05:05.599645826Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -933,7 +933,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:26.996109719Z",
+          "created_at": "2025-08-01T00:05:05.998590959Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -955,7 +955,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:27.358774996Z",
+          "created_at": "2025-08-01T00:05:06.398745325Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -977,7 +977,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:27.731898916Z",
+          "created_at": "2025-08-01T00:05:06.790505624Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -999,7 +999,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:28.096809626Z",
+          "created_at": "2025-08-01T00:05:07.199713609Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1021,7 +1021,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:28.479909418Z",
+          "created_at": "2025-08-01T00:05:07.596500603Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1043,7 +1043,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:28.842369829Z",
+          "created_at": "2025-08-01T00:05:07.997793386Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1065,7 +1065,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:29.219065256Z",
+          "created_at": "2025-08-01T00:05:08.381509773Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1087,7 +1087,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:29.586481232Z",
+          "created_at": "2025-08-01T00:05:08.76579698Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1109,7 +1109,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:29.951118759Z",
+          "created_at": "2025-08-01T00:05:09.159673897Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1131,7 +1131,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:30.341772269Z",
+          "created_at": "2025-08-01T00:05:09.557596611Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1153,7 +1153,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:30.715938769Z",
+          "created_at": "2025-08-01T00:05:09.950543555Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1175,7 +1175,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:31.089904077Z",
+          "created_at": "2025-08-01T00:05:10.351722165Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1197,7 +1197,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:31.464995093Z",
+          "created_at": "2025-08-01T00:05:10.752622361Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1219,7 +1219,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:31.830372387Z",
+          "created_at": "2025-08-01T00:05:11.15541961Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1241,7 +1241,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:32.196652344Z",
+          "created_at": "2025-08-01T00:05:11.549741697Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1263,7 +1263,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:32.564551739Z",
+          "created_at": "2025-08-01T00:05:11.935619908Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1285,7 +1285,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:32.929462678Z",
+          "created_at": "2025-08-01T00:05:12.343367145Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1307,7 +1307,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:33.302119433Z",
+          "created_at": "2025-08-01T00:05:12.745897023Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1329,7 +1329,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:33.664518348Z",
+          "created_at": "2025-08-01T00:05:13.148396264Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1351,7 +1351,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:34.025969567Z",
+          "created_at": "2025-08-01T00:05:13.549096782Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1373,7 +1373,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:34.387819486Z",
+          "created_at": "2025-08-01T00:05:13.945126876Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1395,7 +1395,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:34.755034711Z",
+          "created_at": "2025-08-01T00:05:14.351732762Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1417,7 +1417,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:35.152309014Z",
+          "created_at": "2025-08-01T00:05:14.754792448Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1439,7 +1439,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:35.552405157Z",
+          "created_at": "2025-08-01T00:05:15.157906888Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1461,7 +1461,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:35.928893645Z",
+          "created_at": "2025-08-01T00:05:15.567665265Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1483,7 +1483,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:36.307086641Z",
+          "created_at": "2025-08-01T00:05:15.981925795Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1505,7 +1505,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:36.675663437Z",
+          "created_at": "2025-08-01T00:05:16.388785931Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1527,7 +1527,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:37.047857232Z",
+          "created_at": "2025-08-01T00:05:16.795150512Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1549,7 +1549,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:37.416299637Z",
+          "created_at": "2025-08-01T00:05:17.204509535Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1571,7 +1571,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:37.785786223Z",
+          "created_at": "2025-08-01T00:05:17.613690212Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1593,7 +1593,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:38.143189585Z",
+          "created_at": "2025-08-01T00:05:18.020711094Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1615,7 +1615,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:38.52253153Z",
+          "created_at": "2025-08-01T00:05:18.428597263Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1637,7 +1637,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:38.878479077Z",
+          "created_at": "2025-08-01T00:05:18.836863657Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1659,7 +1659,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:39.236508581Z",
+          "created_at": "2025-08-01T00:05:19.248527489Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1681,7 +1681,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:39.598374943Z",
+          "created_at": "2025-08-01T00:05:19.662063245Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1703,7 +1703,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:39.962586339Z",
+          "created_at": "2025-08-01T00:05:20.074553793Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1725,7 +1725,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:40.353257751Z",
+          "created_at": "2025-08-01T00:05:20.494386446Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1747,7 +1747,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:40.721819096Z",
+          "created_at": "2025-08-01T00:05:20.905809772Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1769,7 +1769,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:41.08041087Z",
+          "created_at": "2025-08-01T00:05:21.32374153Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1791,7 +1791,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:41.442236982Z",
+          "created_at": "2025-08-01T00:05:21.732533121Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1813,7 +1813,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:41.799804132Z",
+          "created_at": "2025-08-01T00:05:22.140888939Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1835,7 +1835,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:42.166527887Z",
+          "created_at": "2025-08-01T00:05:22.552257821Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1857,7 +1857,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:42.527814382Z",
+          "created_at": "2025-08-01T00:05:22.970740344Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1879,7 +1879,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:42.895100966Z",
+          "created_at": "2025-08-01T00:05:23.380926627Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1901,7 +1901,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:43.258686863Z",
+          "created_at": "2025-08-01T00:05:23.790553354Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1923,7 +1923,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:43.619567501Z",
+          "created_at": "2025-08-01T00:05:24.202112923Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1945,7 +1945,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:43.976895586Z",
+          "created_at": "2025-08-01T00:05:24.612103888Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1967,7 +1967,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:44.332362908Z",
+          "created_at": "2025-08-01T00:05:25.019727418Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -1989,7 +1989,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:44.690519899Z",
+          "created_at": "2025-08-01T00:05:25.422980466Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -2011,7 +2011,7 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:45.059572858Z",
+          "created_at": "2025-08-01T00:05:25.815598412Z",
          "done": false,
          "done_reason": null,
          "total_duration": null,
@ -2033,15 +2033,15 @@
        "__type__": "ollama._types.ChatResponse",
        "__data__": {
          "model": "llama3.2-vision:11b",
-          "created_at": "2025-07-31T18:04:45.471945315Z",
+          "created_at": "2025-08-01T00:05:26.224081261Z",
          "done": true,
          "done_reason": "stop",
-          "total_duration": 34274694911,
-          "load_duration": 57572534,
+          "total_duration": 37514337521,
+          "load_duration": 60023634,
          "prompt_eval_count": 18,
-          "prompt_eval_duration": 529292721,
+          "prompt_eval_duration": 561160541,
          "eval_count": 92,
-          "eval_duration": 33679338548,
+          "eval_duration": 36885221241,
          "message": {
            "role": "assistant",
            "content": "",
--- a/tests/integration/vector_io/test_openai_vector_stores.py
+++ b/tests/integration/vector_io/test_openai_vector_stores.py
@ -31,6 +31,7 @@ def skip_if_provider_doesnt_support_openai_vector_stores(client_with_models):
            "remote::chromadb",
            "remote::qdrant",
            "inline::qdrant",
+            "remote::weaviate",
        ]:
            return

@ -111,11 +112,11 @@ def test_openai_create_vector_store(compat_client_with_empty_stores, client_with

    # Create a vector store
    vector_store = client.vector_stores.create(
-        name="test_vector_store", metadata={"purpose": "testing", "environment": "integration"}
+        name="Vs_test_vector_store", metadata={"purpose": "testing", "environment": "integration"}
    )

    assert vector_store is not None
-    assert vector_store.name == "test_vector_store"
+    assert vector_store.name == "Vs_test_vector_store"
    assert vector_store.object == "vector_store"
    assert vector_store.status in ["completed", "in_progress"]
    assert vector_store.metadata["purpose"] == "testing"
--- a/tests/unit/providers/vector_io/test_vector_utils.py
+++ b/tests/unit/providers/vector_io/test_vector_utils.py
@ -5,7 +5,7 @@
 # the root directory of this source tree.

 from llama_stack.apis.vector_io import Chunk, ChunkMetadata
-from llama_stack.providers.utils.vector_io.chunk_utils import generate_chunk_id
+from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id

 # This test is a unit test for the chunk_utils.py helpers. This should only contain
 # tests which are specific to this file. More general (API-level) tests should be placed in
--- a/uv.lock
+++ b/uv.lock