Merge branch 'main' into add-batches

2025-08-15 06:00:48 +00:00 · 2025-08-13 07:33:41 -04:00 · 2025-08-13 07:33:41 -04:00 · 95a3ecdffc
commit 95a3ecdffc
parent 04a73c89ef 6358d0a478
67 changed files with 1158 additions and 424 deletions
--- a/.github/actions/setup-runner/action.yml
+++ b/.github/actions/setup-runner/action.yml
@ -28,7 +28,7 @@ runs:
        # Install llama-stack-client-python based on the client-version input
        if [ "${{ inputs.client-version }}" = "latest" ]; then
          echo "Installing latest llama-stack-client-python from main branch"
-          uv pip install git+https://github.com/meta-llama/llama-stack-client-python.git@main
+          uv pip install git+https://github.com/llamastack/llama-stack-client-python.git@main
        elif [ "${{ inputs.client-version }}" = "published" ]; then
          echo "Installing published llama-stack-client-python from PyPI"
          uv pip install llama-stack-client
--- a/.github/workflows/integration-tests.yml
+++ b/.github/workflows/integration-tests.yml
@ -52,7 +52,8 @@ jobs:
        run: |
          # Get test directories dynamically, excluding non-test directories
          # NOTE: we are excluding post_training since the tests take too long
-          TEST_TYPES=$(find tests/integration -maxdepth 1 -mindepth 1 -type d -printf "%f\n" |
+          TEST_TYPES=$(find tests/integration -maxdepth 1 -mindepth 1 -type d |
+            sed 's|tests/integration/||' |
            grep -Ev "^(__pycache__|fixtures|test_cases|recordings|non_ci|post_training)$" |
            sort | jq -R -s -c 'split("\n")[:-1]')
          echo "test-types=$TEST_TYPES" >> $GITHUB_OUTPUT
--- a/.github/workflows/integration-vector-io-tests.yml
+++ b/.github/workflows/integration-vector-io-tests.yml
@ -164,9 +164,9 @@ jobs:
          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 }}" \
+          uv run pytest -sv --stack-config="files=inline::localfs,inference=inline::sentence-transformers,vector_io=${{ matrix.vector-io-provider }}" \
            tests/integration/vector_io \
-            --embedding-model sentence-transformers/all-MiniLM-L6-v2
+            --embedding-model inline::sentence-transformers/all-MiniLM-L6-v2

      - name: Check Storage and Memory Available After Tests
        if: ${{ always() }}
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@ -1,13 +1,82 @@
-# Contributing to Llama-Stack
+# Contributing to Llama Stack
 We want to make contributing to this project as easy and transparent as
 possible.

+## Set up your development environment
+
+We use [uv](https://github.com/astral-sh/uv) to manage python dependencies and virtual environments.
+You can install `uv` by following this [guide](https://docs.astral.sh/uv/getting-started/installation/).
+
+You can install the dependencies by running:
+
+```bash
+cd llama-stack
+uv sync --group dev
+uv pip install -e .
+source .venv/bin/activate
+```
+
+```{note}
+You can use a specific version of Python with `uv` by adding the `--python <version>` flag (e.g. `--python 3.12`).
+Otherwise, `uv` will automatically select a Python version according to the `requires-python` section of the `pyproject.toml`.
+For more info, see the [uv docs around Python versions](https://docs.astral.sh/uv/concepts/python-versions/).
+```
+
+Note that you can create a dotenv file `.env` that includes necessary environment variables:
+```
+LLAMA_STACK_BASE_URL=http://localhost:8321
+LLAMA_STACK_CLIENT_LOG=debug
+LLAMA_STACK_PORT=8321
+LLAMA_STACK_CONFIG=<provider-name>
+TAVILY_SEARCH_API_KEY=
+BRAVE_SEARCH_API_KEY=
+```
+
+And then use this dotenv file when running client SDK tests via the following:
+```bash
+uv run --env-file .env -- pytest -v tests/integration/inference/test_text_inference.py --text-model=meta-llama/Llama-3.1-8B-Instruct
+```
+
+### Pre-commit Hooks
+
+We use [pre-commit](https://pre-commit.com/) to run linting and formatting checks on your code. You can install the pre-commit hooks by running:
+
+```bash
+uv run pre-commit install
+```
+
+After that, pre-commit hooks will run automatically before each commit.
+
+Alternatively, if you don't want to install the pre-commit hooks, you can run the checks manually by running:
+
+```bash
+uv run pre-commit run --all-files
+```
+
+```{caution}
+Before pushing your changes, make sure that the pre-commit hooks have passed successfully.
+```
+
 ## Discussions -> Issues -> Pull Requests

 We actively welcome your pull requests. However, please read the following. This is heavily inspired by [Ghostty](https://github.com/ghostty-org/ghostty/blob/main/CONTRIBUTING.md).

 If in doubt, please open a [discussion](https://github.com/meta-llama/llama-stack/discussions); we can always convert that to an issue later.

+### Issues
+We use GitHub issues to track public bugs. Please ensure your description is
+clear and has sufficient instructions to be able to reproduce the issue.
+
+Meta has a [bounty program](http://facebook.com/whitehat/info) for the safe
+disclosure of security bugs. In those cases, please go through the process
+outlined on that page and do not file a public issue.
+
+### Contributor License Agreement ("CLA")
+In order to accept your pull request, we need you to submit a CLA. You only need
+to do this once to work on any of Meta's open source projects.
+
+Complete your CLA here: <https://code.facebook.com/cla>
+
 **I'd like to contribute!**

 If you are new to the project, start by looking at the issues tagged with "good first issue". If you're interested
@ -51,93 +120,15 @@ Please avoid picking up too many issues at once. This helps you stay focused and

 Please keep pull requests (PRs) small and focused. If you have a large set of changes, consider splitting them into logically grouped, smaller PRs to facilitate review and testing.

-> [!TIP]
-> As a general guideline:
-> - Experienced contributors should try to keep no more than 5 open PRs at a time.
-> - New contributors are encouraged to have only one open PR at a time until they’re familiar with the codebase and process.
-
-## Contributor License Agreement ("CLA")
-In order to accept your pull request, we need you to submit a CLA. You only need
-to do this once to work on any of Meta's open source projects.
-
-Complete your CLA here: <https://code.facebook.com/cla>
-
-## Issues
-We use GitHub issues to track public bugs. Please ensure your description is
-clear and has sufficient instructions to be able to reproduce the issue.
-
-Meta has a [bounty program](http://facebook.com/whitehat/info) for the safe
-disclosure of security bugs. In those cases, please go through the process
-outlined on that page and do not file a public issue.
-
-
-## Set up your development environment
-
-We use [uv](https://github.com/astral-sh/uv) to manage python dependencies and virtual environments.
-You can install `uv` by following this [guide](https://docs.astral.sh/uv/getting-started/installation/).
-
-You can install the dependencies by running:
-
-```bash
-cd llama-stack
-uv sync --group dev
-uv pip install -e .
-source .venv/bin/activate
+```{tip}
+As a general guideline:
+- Experienced contributors should try to keep no more than 5 open PRs at a time.
+- New contributors are encouraged to have only one open PR at a time until they’re familiar with the codebase and process.
 ```

-> [!NOTE]
-> You can use a specific version of Python with `uv` by adding the `--python <version>` flag (e.g. `--python 3.12`)
-> Otherwise, `uv` will automatically select a Python version according to the `requires-python` section of the `pyproject.toml`.
-> For more info, see the [uv docs around Python versions](https://docs.astral.sh/uv/concepts/python-versions/).
+## Repository guidelines

-Note that you can create a dotenv file `.env` that includes necessary environment variables:
-```
-LLAMA_STACK_BASE_URL=http://localhost:8321
-LLAMA_STACK_CLIENT_LOG=debug
-LLAMA_STACK_PORT=8321
-LLAMA_STACK_CONFIG=<provider-name>
-TAVILY_SEARCH_API_KEY=
-BRAVE_SEARCH_API_KEY=
-```
-
-And then use this dotenv file when running client SDK tests via the following:
-```bash
-uv run --env-file .env -- pytest -v tests/integration/inference/test_text_inference.py --text-model=meta-llama/Llama-3.1-8B-Instruct
-```
-
-## Pre-commit Hooks
-
-We use [pre-commit](https://pre-commit.com/) to run linting and formatting checks on your code. You can install the pre-commit hooks by running:
-
-```bash
-uv run pre-commit install
-```
-
-After that, pre-commit hooks will run automatically before each commit.
-
-Alternatively, if you don't want to install the pre-commit hooks, you can run the checks manually by running:
-
-```bash
-uv run pre-commit run --all-files
-```
-
-> [!CAUTION]
-> Before pushing your changes, make sure that the pre-commit hooks have passed successfully.
-
-## Running tests
-
-You can find the Llama Stack testing documentation [here](https://github.com/meta-llama/llama-stack/blob/main/tests/README.md).
-
-## Adding a new dependency to the project
-
-To add a new dependency to the project, you can use the `uv` command. For example, to add `foo` to the project, you can run:
-
-```bash
-uv add foo
-uv sync
-```
-
-## Coding Style
+### Coding Style

 * Comments should provide meaningful insights into the code. Avoid filler comments that simply
  describe the next step, as they create unnecessary clutter, same goes for docstrings.
@ -159,6 +150,10 @@ uv sync
 * When possible, use keyword arguments only when calling functions.
 * Llama Stack utilizes [custom Exception classes](llama_stack/apis/common/errors.py) for certain Resources that should be used where applicable.

+### License
+By contributing to Llama, you agree that your contributions will be licensed
+under the LICENSE file in the root directory of this source tree.
+
 ## Common Tasks

 Some tips about common tasks you work on while contributing to Llama Stack:
@ -210,8 +205,4 @@ If you modify or add new API endpoints, update the API documentation accordingly
 uv run ./docs/openapi_generator/run_openapi_generator.sh
 ```

-The generated API documentation will be available in `docs/_static/`. Make sure to review the changes before committing.
-
-## License
-By contributing to Llama, you agree that your contributions will be licensed
-under the LICENSE file in the root directory of this source tree.
+The generated API documentation will be available in `docs/_static/`. Make sure to review the changes before committing.
--- a/README.md
+++ b/README.md
@ -1,5 +1,8 @@
 # Llama Stack

+<a href="https://trendshift.io/repositories/11824" target="_blank"><img src="https://trendshift.io/api/badge/repositories/11824" alt="meta-llama%2Fllama-stack | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+
+-----
 [![PyPI version](https://img.shields.io/pypi/v/llama_stack.svg)](https://pypi.org/project/llama_stack/)
 [![PyPI - Downloads](https://img.shields.io/pypi/dm/llama-stack)](https://pypi.org/project/llama-stack/)
 [![License](https://img.shields.io/pypi/l/llama_stack.svg)](https://github.com/meta-llama/llama-stack/blob/main/LICENSE)
@ -9,6 +12,7 @@

 [**Quick Start**](https://llama-stack.readthedocs.io/en/latest/getting_started/index.html) | [**Documentation**](https://llama-stack.readthedocs.io/en/latest/index.html) | [**Colab Notebook**](./docs/getting_started.ipynb) | [**Discord**](https://discord.gg/llama-stack)

+
 ### ✨🎉 Llama 4 Support  🎉✨
 We released [Version 0.2.0](https://github.com/meta-llama/llama-stack/releases/tag/v0.2.0) with support for the Llama 4 herd of models released by Meta.

@ -179,3 +183,17 @@ Please checkout our [Documentation](https://llama-stack.readthedocs.io/en/latest
 Check out our client SDKs for connecting to a Llama Stack server in your preferred language, you can choose from [python](https://github.com/meta-llama/llama-stack-client-python), [typescript](https://github.com/meta-llama/llama-stack-client-typescript), [swift](https://github.com/meta-llama/llama-stack-client-swift), and [kotlin](https://github.com/meta-llama/llama-stack-client-kotlin) programming languages to quickly build your applications.

 You can find more example scripts with client SDKs to talk with the Llama Stack server in our [llama-stack-apps](https://github.com/meta-llama/llama-stack-apps/tree/main/examples) repo.
+
+
+## 🌟 GitHub Star History
+## Star History
+
+[![Star History Chart](https://api.star-history.com/svg?repos=meta-llama/llama-stack&type=Date)](https://www.star-history.com/#meta-llama/llama-stack&Date)
+
+## ✨ Contributors
+
+Thanks to all of our amazing contributors!
+
+<a href="https://github.com/meta-llama/llama-stack/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=meta-llama/llama-stack" />
+</a>
--- a/docs/_static/js/keyboard_shortcuts.js
+++ b/docs/_static/js/keyboard_shortcuts.js
@ -0,0 +1,14 @@
+document.addEventListener('keydown', function(event) {
+  // command+K or ctrl+K
+  if ((event.metaKey || event.ctrlKey) && event.key === 'k') {
+    event.preventDefault();
+    document.querySelector('.search-input, .search-field, input[name="q"]').focus();
+  }
+
+  // forward slash
+  if (event.key === '/' &&
+      !event.target.matches('input, textarea, select')) {
+    event.preventDefault();
+    document.querySelector('.search-input, .search-field, input[name="q"]').focus();
+  }
+});
--- a/docs/_static/llama-stack-spec.html
+++ b/docs/_static/llama-stack-spec.html
@ -8293,28 +8293,60 @@
                        "type": "array",
                        "items": {
                            "type": "object",
-                            "additionalProperties": {
-                                "oneOf": [
-                                    {
-                                        "type": "null"
+                            "properties": {
+                                "attributes": {
+                                    "type": "object",
+                                    "additionalProperties": {
+                                        "oneOf": [
+                                            {
+                                                "type": "null"
+                                            },
+                                            {
+                                                "type": "boolean"
+                                            },
+                                            {
+                                                "type": "number"
+                                            },
+                                            {
+                                                "type": "string"
+                                            },
+                                            {
+                                                "type": "array"
+                                            },
+                                            {
+                                                "type": "object"
+                                            }
+                                        ]
                                    },
-                                    {
-                                        "type": "boolean"
-                                    },
-                                    {
-                                        "type": "number"
-                                    },
-                                    {
-                                        "type": "string"
-                                    },
-                                    {
-                                        "type": "array"
-                                    },
-                                    {
-                                        "type": "object"
-                                    }
-                                ]
-                            }
+                                    "description": "(Optional) Key-value attributes associated with the file"
+                                },
+                                "file_id": {
+                                    "type": "string",
+                                    "description": "Unique identifier of the file containing the result"
+                                },
+                                "filename": {
+                                    "type": "string",
+                                    "description": "Name of the file containing the result"
+                                },
+                                "score": {
+                                    "type": "number",
+                                    "description": "Relevance score for this search result (between 0 and 1)"
+                                },
+                                "text": {
+                                    "type": "string",
+                                    "description": "Text content of the search result"
+                                }
+                            },
+                            "additionalProperties": false,
+                            "required": [
+                                "attributes",
+                                "file_id",
+                                "filename",
+                                "score",
+                                "text"
+                            ],
+                            "title": "OpenAIResponseOutputMessageFileSearchToolCallResults",
+                            "description": "Search results returned by the file search operation."
                        },
                        "description": "(Optional) Search results returned by the file search operation"
                    }
@ -8515,6 +8547,13 @@
                            "$ref": "#/components/schemas/OpenAIResponseInputTool"
                        }
                    },
+                    "include": {
+                        "type": "array",
+                        "items": {
+                            "type": "string"
+                        },
+                        "description": "(Optional) Additional fields to include in the response."
+                    },
                    "max_infer_iters": {
                        "type": "integer"
                    }
--- a/docs/_static/llama-stack-spec.yaml
+++ b/docs/_static/llama-stack-spec.yaml
@ -6021,14 +6021,44 @@ components:
          type: array
          items:
            type: object
-            additionalProperties:
-              oneOf:
-                - type: 'null'
-                - type: boolean
-                - type: number
-                - type: string
-                - type: array
-                - type: object
+            properties:
+              attributes:
+                type: object
+                additionalProperties:
+                  oneOf:
+                    - type: 'null'
+                    - type: boolean
+                    - type: number
+                    - type: string
+                    - type: array
+                    - type: object
+                description: >-
+                  (Optional) Key-value attributes associated with the file
+              file_id:
+                type: string
+                description: >-
+                  Unique identifier of the file containing the result
+              filename:
+                type: string
+                description: Name of the file containing the result
+              score:
+                type: number
+                description: >-
+                  Relevance score for this search result (between 0 and 1)
+              text:
+                type: string
+                description: Text content of the search result
+            additionalProperties: false
+            required:
+              - attributes
+              - file_id
+              - filename
+              - score
+              - text
+            title: >-
+              OpenAIResponseOutputMessageFileSearchToolCallResults
+            description: >-
+              Search results returned by the file search operation.
          description: >-
            (Optional) Search results returned by the file search operation
      additionalProperties: false
@ -6188,6 +6218,12 @@ components:
          type: array
          items:
            $ref: '#/components/schemas/OpenAIResponseInputTool'
+        include:
+          type: array
+          items:
+            type: string
+          description: >-
+            (Optional) Additional fields to include in the response.
        max_infer_iters:
          type: integer
      additionalProperties: false
--- a/docs/source/apis/external.md
+++ b/docs/source/apis/external.md
@ -111,7 +111,7 @@ name = "llama-stack-api-weather"
 version = "0.1.0"
 description = "Weather API for Llama Stack"
 readme = "README.md"
-requires-python = ">=3.10"
+requires-python = ">=3.12"
 dependencies = ["llama-stack", "pydantic"]

 [build-system]
@ -231,7 +231,7 @@ name = "llama-stack-provider-kaze"
 version = "0.1.0"
 description = "Kaze weather provider for Llama Stack"
 readme = "README.md"
-requires-python = ">=3.10"
+requires-python = ">=3.12"
 dependencies = ["llama-stack", "pydantic", "aiohttp"]

 [build-system]
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@ -131,6 +131,7 @@ html_static_path = ["../_static"]
 def setup(app):
    app.add_css_file("css/my_theme.css")
    app.add_js_file("js/detect_theme.js")
+    app.add_js_file("js/keyboard_shortcuts.js")

    def dockerhub_role(name, rawtext, text, lineno, inliner, options={}, content=[]):
        url = f"https://hub.docker.com/r/llamastack/{text}"
--- a/docs/source/contributing/index.md
+++ b/docs/source/contributing/index.md
@ -2,14 +2,28 @@
 ```{include} ../../../CONTRIBUTING.md
 ```

-See the [Adding a New API Provider](new_api_provider.md) which describes how to add new API providers to the Stack.
+## Adding a New Provider

+See the [Adding a New API Provider Page](new_api_provider.md) which describes how to add new API providers to the Stack.

+See the [Vector Database Page](new_vector_database.md) which describes how to add a new vector databases with Llama Stack.

+See the [External Provider Page](../providers/external/index.md) which describes how to add external providers to the Stack.
 ```{toctree}
 :maxdepth: 1
 :hidden:

 new_api_provider
-testing
+new_vector_database
 ```
+
+## Testing
+
+See the [Test Page](testing.md) which describes how to test your changes.
+```{toctree}
+:maxdepth: 1
+:hidden:
+:caption: Testing
+
+testing
+```
--- a/docs/source/contributing/new_vector_database.md
+++ b/docs/source/contributing/new_vector_database.md
@ -0,0 +1,75 @@
+# Adding a New Vector Database
+
+This guide will walk you through the process of adding a new vector database to Llama Stack.
+
+> **_NOTE:_** Here's an example Pull Request of the [Milvus Vector Database Provider](https://github.com/meta-llama/llama-stack/pull/1467).
+
+Vector Database providers are used to store and retrieve vector embeddings. Vector databases are not limited to vector
+search but can support keyword and hybrid search. Additionally, vector database can also support operations like
+filtering, sorting, and aggregating vectors.
+
+## Steps to Add a New Vector Database Provider
+1. **Choose the Database Type**: Determine if your vector database is a remote service, inline, or both.
+   - Remote databases make requests to external services, while inline databases execute locally. Some providers support both.
+2. **Implement the Provider**: Create a new provider class that inherits from `VectorDatabaseProvider` and implements the required methods.
+   - Implement methods for vector storage, retrieval, search, and any additional features your database supports.
+     - You will need to implement the following methods for `YourVectorIndex`:
+        - `YourVectorIndex.create()`
+        - `YourVectorIndex.initialize()`
+        - `YourVectorIndex.add_chunks()`
+        - `YourVectorIndex.delete_chunk()`
+        - `YourVectorIndex.query_vector()`
+        - `YourVectorIndex.query_keyword()`
+        - `YourVectorIndex.query_hybrid()`
+     - You will need to implement the following methods for `YourVectorIOAdapter`:
+        - `YourVectorIOAdapter.initialize()`
+        - `YourVectorIOAdapter.shutdown()`
+        - `YourVectorIOAdapter.list_vector_dbs()`
+        - `YourVectorIOAdapter.register_vector_db()`
+        - `YourVectorIOAdapter.unregister_vector_db()`
+        - `YourVectorIOAdapter.insert_chunks()`
+        - `YourVectorIOAdapter.query_chunks()`
+        - `YourVectorIOAdapter.delete_chunks()`
+3. **Add to Registry**: Register your provider in the appropriate registry file.
+   - Update {repopath}`llama_stack/providers/registry/vector_io.py` to include your new provider.
+```python
+from llama_stack.providers.registry.specs import InlineProviderSpec
+from llama_stack.providers.registry.api import Api
+
+InlineProviderSpec(
+    api=Api.vector_io,
+    provider_type="inline::milvus",
+    pip_packages=["pymilvus>=2.4.10"],
+    module="llama_stack.providers.inline.vector_io.milvus",
+    config_class="llama_stack.providers.inline.vector_io.milvus.MilvusVectorIOConfig",
+    api_dependencies=[Api.inference],
+    optional_api_dependencies=[Api.files],
+    description="",
+),
+```
+4. **Add Tests**: Create unit tests and integration tests for your provider in the `tests/` directory.
+   - Unit Tests
+     - By following the structure of the class methods, you will be able to easily run unit and integration tests for your database.
+       1. You have to configure the tests for your provide in `/tests/unit/providers/vector_io/conftest.py`.
+       2. Update the `vector_provider` fixture to include your provider if they are an inline provider.
+       3. Create a `your_vectorprovider_index` fixture that initializes your vector index.
+       4. Create a `your_vectorprovider_adapter` fixture that initializes your vector adapter.
+       5. Add your provider to the `vector_io_providers` fixture dictionary.
+         - Please follow the naming convention of `your_vectorprovider_index` and `your_vectorprovider_adapter` as the tests require this to execute properly.
+   - Integration Tests
+     - Integration tests are located in {repopath}`tests/integration`. These tests use the python client-SDK APIs (from the `llama_stack_client` package) to test functionality.
+     - The two set of integration tests are:
+       - `tests/integration/vector_io/test_vector_io.py`: This file tests registration, insertion, and retrieval.
+       - `tests/integration/vector_io/test_openai_vector_stores.py`: These tests are for OpenAI-compatible vector stores and test the OpenAI API compatibility.
+        - You will need to update `skip_if_provider_doesnt_support_openai_vector_stores` to include your provider as well as `skip_if_provider_doesnt_support_openai_vector_stores_search` to test the appropriate search functionality.
+     - Running the tests in the GitHub CI
+       - You will need to update the `.github/workflows/integration-vector-io-tests.yml` file to include your provider.
+        - If your provider is a remote provider, you will also have to add a container to spin up and run it in the action.
+   - Updating the pyproject.yml
+     - If you are adding tests for the `inline` provider you will have to update the `unit` group.
+       - `uv add new_pip_package --group unit`
+     - If you are adding tests for the `remote` provider you will have to update the `test` group, which is used in the GitHub CI for integration tests.
+       - `uv add new_pip_package --group test`
+5. **Update Documentation**: Please update the documentation for end users
+   - Generate the provider documentation by running {repopath}`./scripts/provider_codegen.py`.
+   - Update the autogenerated content in the registry/vector_io.py file with information about your provider. Please see other providers for examples.
--- a/docs/source/contributing/testing.md
+++ b/docs/source/contributing/testing.md
@ -1,6 +1,8 @@
-# Testing Llama Stack
+```{include} ../../../tests/README.md
+```

-Tests are of three different kinds:
- Unit tests
- Provider focused integration tests
- Client SDK tests
+```{include} ../../../tests/unit/README.md
+```
+
+```{include} ../../../tests/integration/README.md
+```
--- a/docs/source/providers/external/external-providers-guide.md
+++ b/docs/source/providers/external/external-providers-guide.md
@ -226,7 +226,7 @@ uv init
 name = "llama-stack-provider-ollama"
 version = "0.1.0"
 description = "Ollama provider for Llama Stack"
-requires-python = ">=3.10"
+requires-python = ">=3.12"
 dependencies = ["llama-stack", "pydantic", "ollama", "aiohttp"]
 ```

--- a/docs/source/providers/inference/index.md
+++ b/docs/source/providers/inference/index.md
@ -35,6 +35,7 @@ remote_runpod
 remote_sambanova
 remote_tgi
 remote_together
+remote_vertexai
 remote_vllm
 remote_watsonx
 ```
--- a/docs/source/providers/inference/remote_vertexai.md
+++ b/docs/source/providers/inference/remote_vertexai.md
@ -0,0 +1,40 @@
+# remote::vertexai
+
+## Description
+
+Google Vertex AI inference provider enables you to use Google's Gemini models through Google Cloud's Vertex AI platform, providing several advantages:
+
+• Enterprise-grade security: Uses Google Cloud's security controls and IAM
+• Better integration: Seamless integration with other Google Cloud services
+• Advanced features: Access to additional Vertex AI features like model tuning and monitoring
+• Authentication: Uses Google Cloud Application Default Credentials (ADC) instead of API keys
+
+Configuration:
+- Set VERTEX_AI_PROJECT environment variable (required)
+- Set VERTEX_AI_LOCATION environment variable (optional, defaults to us-central1)
+- Use Google Cloud Application Default Credentials or service account key
+
+Authentication Setup:
+Option 1 (Recommended): gcloud auth application-default login
+Option 2: Set GOOGLE_APPLICATION_CREDENTIALS to service account key path
+
+Available Models:
+- vertex_ai/gemini-2.0-flash
+- vertex_ai/gemini-2.5-flash
+- vertex_ai/gemini-2.5-pro
+
+## Configuration
+
+| Field | Type | Required | Default | Description |
+|-------|------|----------|---------|-------------|
+| `project` | `<class 'str'>` | No |  | Google Cloud project ID for Vertex AI |
+| `location` | `<class 'str'>` | No | us-central1 | Google Cloud location for Vertex AI |
+
+## Sample Configuration
+
+```yaml
+project: ${env.VERTEX_AI_PROJECT:=}
+location: ${env.VERTEX_AI_LOCATION:=us-central1}
+
+```
+
--- a/docs/source/providers/vector_io/inline_faiss.md
+++ b/docs/source/providers/vector_io/inline_faiss.md
@ -12,6 +12,18 @@ That means you'll get fast and efficient vector retrieval.
 - Lightweight and easy to use
 - Fully integrated with Llama Stack
 - GPU support
+- **Vector search** - FAISS supports pure vector similarity search using embeddings
+
+## Search Modes
+
+**Supported:**
+- **Vector Search** (`mode="vector"`): Performs vector similarity search using embeddings
+
+**Not Supported:**
+- **Keyword Search** (`mode="keyword"`): Not supported by FAISS
+- **Hybrid Search** (`mode="hybrid"`): Not supported by FAISS
+
+> **Note**: FAISS is designed as a pure vector similarity search library. See the [FAISS GitHub repository](https://github.com/facebookresearch/faiss) for more details about FAISS's core functionality.

 ## Usage

--- a/docs/source/providers/vector_io/remote_milvus.md
+++ b/docs/source/providers/vector_io/remote_milvus.md
@ -11,6 +11,7 @@ That means you're not limited to storing vectors in memory or in a separate serv

 - Easy to use
 - Fully integrated with Llama Stack
+- Supports all search modes: vector, keyword, and hybrid search (both inline and remote configurations)

 ## Usage

@ -101,6 +102,92 @@ vector_io:
 - **`client_pem_path`**: Path to the **client certificate** file (required for mTLS).
 - **`client_key_path`**: Path to the **client private key** file (required for mTLS).

+## Search Modes
+
+Milvus supports three different search modes for both inline and remote configurations:
+
+### Vector Search
+Vector search uses semantic similarity to find the most relevant chunks based on embedding vectors. This is the default search mode and works well for finding conceptually similar content.
+
+```python
+# Vector search example
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="What is machine learning?",
+    search_mode="vector",
+    max_num_results=5,
+)
+```
+
+### Keyword Search
+Keyword search uses traditional text-based matching to find chunks containing specific terms or phrases. This is useful when you need exact term matches.
+
+```python
+# Keyword search example
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="Python programming language",
+    search_mode="keyword",
+    max_num_results=5,
+)
+```
+
+### Hybrid Search
+Hybrid search combines both vector and keyword search methods to provide more comprehensive results. It leverages the strengths of both semantic similarity and exact term matching.
+
+#### Basic Hybrid Search
+```python
+# Basic hybrid search example (uses RRF ranker with default impact_factor=60.0)
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+)
+```
+
+**Note**: The default `impact_factor` value of 60.0 was empirically determined to be optimal in the original RRF research paper: ["Reciprocal Rank Fusion outperforms Condorcet and individual Rank Learning Methods"](https://plg.uwaterloo.ca/~gvcormac/cormacksigir09-rrf.pdf) (Cormack et al., 2009).
+
+#### Hybrid Search with RRF (Reciprocal Rank Fusion) Ranker
+RRF combines rankings from vector and keyword search by using reciprocal ranks. The impact factor controls how much weight is given to higher-ranked results.
+
+```python
+# Hybrid search with custom RRF parameters
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+    ranking_options={
+        "ranker": {
+            "type": "rrf",
+            "impact_factor": 100.0,  # Higher values give more weight to top-ranked results
+        }
+    },
+)
+```
+
+#### Hybrid Search with Weighted Ranker
+Weighted ranker linearly combines normalized scores from vector and keyword search. The alpha parameter controls the balance between the two search methods.
+
+```python
+# Hybrid search with weighted ranker
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+    ranking_options={
+        "ranker": {
+            "type": "weighted",
+            "alpha": 0.7,  # 70% vector search, 30% keyword search
+        }
+    },
+)
+```
+
+For detailed documentation on RRF and Weighted rankers, please refer to the [Milvus Reranking Guide](https://milvus.io/docs/reranking.md).
+
 ## Documentation
 See the [Milvus documentation](https://milvus.io/docs/install-overview.md) for more details about Milvus in general.

--- a/llama_stack/apis/agents/agents.py
+++ b/llama_stack/apis/agents/agents.py
@ -706,6 +706,7 @@ class Agents(Protocol):
        temperature: float | None = None,
        text: OpenAIResponseText | None = None,
        tools: list[OpenAIResponseInputTool] | None = None,
+        include: list[str] | None = None,
        max_infer_iters: int | None = 10,  # this is an extension to the OpenAI API
    ) -> OpenAIResponseObject | AsyncIterator[OpenAIResponseObjectStream]:
        """Create a new OpenAI response.
@ -713,6 +714,7 @@ class Agents(Protocol):
        :param input: Input message(s) to create the response.
        :param model: The underlying LLM used for completions.
        :param previous_response_id: (Optional) if specified, the new response will be a continuation of the previous response. This can be used to easily fork-off new responses from existing responses.
+        :param include: (Optional) Additional fields to include in the response.
        :returns: An OpenAIResponseObject.
        """
        ...
--- a/llama_stack/apis/agents/openai_responses.py
+++ b/llama_stack/apis/agents/openai_responses.py
@ -170,6 +170,23 @@ class OpenAIResponseOutputMessageWebSearchToolCall(BaseModel):
    type: Literal["web_search_call"] = "web_search_call"


+class OpenAIResponseOutputMessageFileSearchToolCallResults(BaseModel):
+    """Search results returned by the file search operation.
+
+    :param attributes: (Optional) Key-value attributes associated with the file
+    :param file_id: Unique identifier of the file containing the result
+    :param filename: Name of the file containing the result
+    :param score: Relevance score for this search result (between 0 and 1)
+    :param text: Text content of the search result
+    """
+
+    attributes: dict[str, Any]
+    file_id: str
+    filename: str
+    score: float
+    text: str
+
+
@json_schema_type
 class OpenAIResponseOutputMessageFileSearchToolCall(BaseModel):
    """File search tool call output message for OpenAI responses.
@ -185,7 +202,7 @@ class OpenAIResponseOutputMessageFileSearchToolCall(BaseModel):
    queries: list[str]
    status: str
    type: Literal["file_search_call"] = "file_search_call"
-    results: list[dict[str, Any]] | None = None
+    results: list[OpenAIResponseOutputMessageFileSearchToolCallResults] | None = None


@json_schema_type
--- a/llama_stack/apis/common/errors.py
+++ b/llama_stack/apis/common/errors.py
@ -67,5 +67,14 @@ class SessionNotFoundError(ValueError):
 class ConflictError(ValueError):
    """raised when an operation cannot be performed due to a conflict with the current state"""

-    def __init__(self, message: str) -> None:
+    pass
+
+
+class ModelTypeError(TypeError):
+    """raised when a model is present but not the correct type"""
+
+    def __init__(self, model_name: str, model_type: str, expected_model_type: str) -> None:
+        message = (
+            f"Model '{model_name}' is of type '{model_type}' rather than the expected type '{expected_model_type}'"
+        )
        super().__init__(message)
--- a/llama_stack/core/build.py
+++ b/llama_stack/core/build.py
@ -91,7 +91,7 @@ def get_provider_dependencies(


 def print_pip_install_help(config: BuildConfig):
-    normal_deps, special_deps = get_provider_dependencies(config)
+    normal_deps, special_deps, _ = get_provider_dependencies(config)

    cprint(
        f"Please install needed dependencies using the following commands:\n\nuv pip install {' '.join(normal_deps)}",
--- a/llama_stack/core/routers/inference.py
+++ b/llama_stack/core/routers/inference.py
@ -18,7 +18,7 @@ from llama_stack.apis.common.content_types import (
    InterleavedContent,
    InterleavedContentItem,
 )
-from llama_stack.apis.common.errors import ModelNotFoundError
+from llama_stack.apis.common.errors import ModelNotFoundError, ModelTypeError
 from llama_stack.apis.inference import (
    BatchChatCompletionResponse,
    BatchCompletionResponse,
@ -65,7 +65,7 @@ from llama_stack.providers.datatypes import HealthResponse, HealthStatus, Routin
 from llama_stack.providers.utils.inference.inference_store import InferenceStore
 from llama_stack.providers.utils.telemetry.tracing import get_current_span

-logger = get_logger(name=__name__, category="core")
+logger = get_logger(name=__name__, category="inference")


 class InferenceRouter(Inference):
@ -177,6 +177,15 @@ class InferenceRouter(Inference):
            encoded = self.formatter.encode_content(messages)
        return len(encoded.tokens) if encoded and encoded.tokens else 0

+    async def _get_model(self, model_id: str, expected_model_type: str) -> Model:
+        """takes a model id and gets model after ensuring that it is accessible and of the correct type"""
+        model = await self.routing_table.get_model(model_id)
+        if model is None:
+            raise ModelNotFoundError(model_id)
+        if model.model_type != expected_model_type:
+            raise ModelTypeError(model_id, model.model_type, expected_model_type)
+        return model
+
    async def chat_completion(
        self,
        model_id: str,
@ -195,11 +204,7 @@ class InferenceRouter(Inference):
        )
        if sampling_params is None:
            sampling_params = SamplingParams()
-        model = await self.routing_table.get_model(model_id)
-        if model is None:
-            raise ModelNotFoundError(model_id)
-        if model.model_type == ModelType.embedding:
-            raise ValueError(f"Model '{model_id}' is an embedding model and does not support chat completions")
+        model = await self._get_model(model_id, ModelType.llm)
        if tool_config:
            if tool_choice and tool_choice != tool_config.tool_choice:
                raise ValueError("tool_choice and tool_config.tool_choice must match")
@ -301,11 +306,7 @@ class InferenceRouter(Inference):
        logger.debug(
            f"InferenceRouter.completion: {model_id=}, {stream=}, {content=}, {sampling_params=}, {response_format=}",
        )
-        model = await self.routing_table.get_model(model_id)
-        if model is None:
-            raise ModelNotFoundError(model_id)
-        if model.model_type == ModelType.embedding:
-            raise ValueError(f"Model '{model_id}' is an embedding model and does not support chat completions")
+        model = await self._get_model(model_id, ModelType.llm)
        provider = await self.routing_table.get_provider_impl(model_id)
        params = dict(
            model_id=model_id,
@ -355,11 +356,7 @@ class InferenceRouter(Inference):
        task_type: EmbeddingTaskType | None = None,
    ) -> EmbeddingsResponse:
        logger.debug(f"InferenceRouter.embeddings: {model_id}")
-        model = await self.routing_table.get_model(model_id)
-        if model is None:
-            raise ModelNotFoundError(model_id)
-        if model.model_type == ModelType.llm:
-            raise ValueError(f"Model '{model_id}' is an LLM model and does not support embeddings")
+        await self._get_model(model_id, ModelType.embedding)
        provider = await self.routing_table.get_provider_impl(model_id)
        return await provider.embeddings(
            model_id=model_id,
@ -395,12 +392,7 @@ class InferenceRouter(Inference):
        logger.debug(
            f"InferenceRouter.openai_completion: {model=}, {stream=}, {prompt=}",
        )
-        model_obj = await self.routing_table.get_model(model)
-        if model_obj is None:
-            raise ModelNotFoundError(model)
-        if model_obj.model_type == ModelType.embedding:
-            raise ValueError(f"Model '{model}' is an embedding model and does not support completions")
-
+        model_obj = await self._get_model(model, ModelType.llm)
        params = dict(
            model=model_obj.identifier,
            prompt=prompt,
@ -476,11 +468,7 @@ class InferenceRouter(Inference):
        logger.debug(
            f"InferenceRouter.openai_chat_completion: {model=}, {stream=}, {messages=}",
        )
-        model_obj = await self.routing_table.get_model(model)
-        if model_obj is None:
-            raise ModelNotFoundError(model)
-        if model_obj.model_type == ModelType.embedding:
-            raise ValueError(f"Model '{model}' is an embedding model and does not support chat completions")
+        model_obj = await self._get_model(model, ModelType.llm)

        # Use the OpenAI client for a bit of extra input validation without
        # exposing the OpenAI client itself as part of our API surface
@ -567,12 +555,7 @@ class InferenceRouter(Inference):
        logger.debug(
            f"InferenceRouter.openai_embeddings: {model=}, input_type={type(input)}, {encoding_format=}, {dimensions=}",
        )
-        model_obj = await self.routing_table.get_model(model)
-        if model_obj is None:
-            raise ModelNotFoundError(model)
-        if model_obj.model_type != ModelType.embedding:
-            raise ValueError(f"Model '{model}' is not an embedding model")
-
+        model_obj = await self._get_model(model, ModelType.embedding)
        params = dict(
            model=model_obj.identifier,
            input=input,
@ -871,4 +854,5 @@ class InferenceRouter(Inference):
                    model=model.identifier,
                    object="chat.completion",
                )
+                logger.debug(f"InferenceRouter.completion_response: {final_response}")
                await self.store.store_chat_completion(final_response, messages)
--- a/llama_stack/core/routing_tables/models.py
+++ b/llama_stack/core/routing_tables/models.py
@ -63,6 +63,8 @@ class ModelsRoutingTable(CommonRoutingTableImpl, Models):

    async def get_provider_impl(self, model_id: str) -> Any:
        model = await lookup_model(self, model_id)
+        if model.provider_id not in self.impls_by_provider_id:
+            raise ValueError(f"Provider {model.provider_id} not found in the routing table")
        return self.impls_by_provider_id[model.provider_id]

    async def register_model(
--- a/llama_stack/core/routing_tables/toolgroups.py
+++ b/llama_stack/core/routing_tables/toolgroups.py
@ -124,10 +124,7 @@ class ToolGroupsRoutingTable(CommonRoutingTableImpl, ToolGroups):
        return toolgroup

    async def unregister_toolgroup(self, toolgroup_id: str) -> None:
-        tool_group = await self.get_tool_group(toolgroup_id)
-        if tool_group is None:
-            raise ToolGroupNotFoundError(toolgroup_id)
-        await self.unregister_object(tool_group)
+        await self.unregister_object(await self.get_tool_group(toolgroup_id))

    async def shutdown(self) -> None:
        pass
--- a/llama_stack/core/routing_tables/vector_dbs.py
+++ b/llama_stack/core/routing_tables/vector_dbs.py
@ -8,7 +8,7 @@ from typing import Any

 from pydantic import TypeAdapter

-from llama_stack.apis.common.errors import ModelNotFoundError, VectorStoreNotFoundError
+from llama_stack.apis.common.errors import ModelNotFoundError, ModelTypeError, VectorStoreNotFoundError
 from llama_stack.apis.models import ModelType
 from llama_stack.apis.resource import ResourceType
 from llama_stack.apis.vector_dbs import ListVectorDBsResponse, VectorDB, VectorDBs
@ -66,7 +66,7 @@ class VectorDBsRoutingTable(CommonRoutingTableImpl, VectorDBs):
        if model is None:
            raise ModelNotFoundError(embedding_model)
        if model.model_type != ModelType.embedding:
-            raise ValueError(f"Model {embedding_model} is not an embedding model")
+            raise ModelTypeError(embedding_model, model.model_type, ModelType.embedding)
        if "embedding_dimension" not in model.metadata:
            raise ValueError(f"Model {embedding_model} does not have an embedding dimension")
        vector_db_data = {
--- a/llama_stack/distributions/ci-tests/build.yaml
+++ b/llama_stack/distributions/ci-tests/build.yaml
@ -14,6 +14,7 @@ distribution_spec:
    - provider_type: remote::openai
    - provider_type: remote::anthropic
    - provider_type: remote::gemini
+    - provider_type: remote::vertexai
    - provider_type: remote::groq
    - provider_type: remote::sambanova
    - provider_type: inline::sentence-transformers
--- a/llama_stack/distributions/ci-tests/run.yaml
+++ b/llama_stack/distributions/ci-tests/run.yaml
@ -65,6 +65,11 @@ providers:
    provider_type: remote::gemini
    config:
      api_key: ${env.GEMINI_API_KEY:=}
+  - provider_id: ${env.VERTEX_AI_PROJECT:+vertexai}
+    provider_type: remote::vertexai
+    config:
+      project: ${env.VERTEX_AI_PROJECT:=}
+      location: ${env.VERTEX_AI_LOCATION:=us-central1}
  - provider_id: groq
    provider_type: remote::groq
    config:
--- a/llama_stack/distributions/starter/build.yaml
+++ b/llama_stack/distributions/starter/build.yaml
@ -14,6 +14,7 @@ distribution_spec:
    - provider_type: remote::openai
    - provider_type: remote::anthropic
    - provider_type: remote::gemini
+    - provider_type: remote::vertexai
    - provider_type: remote::groq
    - provider_type: remote::sambanova
    - provider_type: inline::sentence-transformers
--- a/llama_stack/distributions/starter/run.yaml
+++ b/llama_stack/distributions/starter/run.yaml
@ -65,6 +65,11 @@ providers:
    provider_type: remote::gemini
    config:
      api_key: ${env.GEMINI_API_KEY:=}
+  - provider_id: ${env.VERTEX_AI_PROJECT:+vertexai}
+    provider_type: remote::vertexai
+    config:
+      project: ${env.VERTEX_AI_PROJECT:=}
+      location: ${env.VERTEX_AI_LOCATION:=us-central1}
  - provider_id: groq
    provider_type: remote::groq
    config:
--- a/llama_stack/distributions/starter/starter.py
+++ b/llama_stack/distributions/starter/starter.py
@ -56,6 +56,7 @@ ENABLED_INFERENCE_PROVIDERS = [
    "fireworks",
    "together",
    "gemini",
+    "vertexai",
    "groq",
    "sambanova",
    "anthropic",
@ -71,6 +72,7 @@ INFERENCE_PROVIDER_IDS = {
    "tgi": "${env.TGI_URL:+tgi}",
    "cerebras": "${env.CEREBRAS_API_KEY:+cerebras}",
    "nvidia": "${env.NVIDIA_API_KEY:+nvidia}",
+    "vertexai": "${env.VERTEX_AI_PROJECT:+vertexai}",
 }


@ -246,6 +248,14 @@ def get_distribution_template() -> DistributionTemplate:
                "",
                "Gemini API Key",
            ),
+            "VERTEX_AI_PROJECT": (
+                "",
+                "Google Cloud Project ID for Vertex AI",
+            ),
+            "VERTEX_AI_LOCATION": (
+                "us-central1",
+                "Google Cloud Location for Vertex AI",
+            ),
            "SAMBANOVA_API_KEY": (
                "",
                "SambaNova API Key",
--- a/llama_stack/log.py
+++ b/llama_stack/log.py
@ -32,6 +32,7 @@ CATEGORIES = [
    "tools",
    "client",
    "telemetry",
+    "openai_responses",
 ]

 # Initialize category levels with default level
@ -99,7 +100,8 @@ def parse_environment_config(env_config: str) -> dict[str, int]:
        Dict[str, int]: A dictionary mapping categories to their log levels.
    """
    category_levels = {}
-    for pair in env_config.split(";"):
+    delimiter = ","
+    for pair in env_config.split(delimiter):
        if not pair.strip():
            continue

--- a/llama_stack/models/llama/llama3/chat_format.py
+++ b/llama_stack/models/llama/llama3/chat_format.py
@ -236,6 +236,7 @@ class ChatFormat:
                    arguments_json=json.dumps(tool_arguments),
                )
            )
+            content = ""

        return RawMessage(
            role="assistant",
--- a/llama_stack/providers/inline/agents/meta_reference/agents.py
+++ b/llama_stack/providers/inline/agents/meta_reference/agents.py
@ -327,10 +327,21 @@ class MetaReferenceAgentsImpl(Agents):
        temperature: float | None = None,
        text: OpenAIResponseText | None = None,
        tools: list[OpenAIResponseInputTool] | None = None,
+        include: list[str] | None = None,
        max_infer_iters: int | None = 10,
    ) -> OpenAIResponseObject:
        return await self.openai_responses_impl.create_openai_response(
-            input, model, instructions, previous_response_id, store, stream, temperature, text, tools, max_infer_iters
+            input,
+            model,
+            instructions,
+            previous_response_id,
+            store,
+            stream,
+            temperature,
+            text,
+            tools,
+            include,
+            max_infer_iters,
        )

    async def list_openai_responses(
--- a/llama_stack/providers/inline/agents/meta_reference/openai_responses.py
+++ b/llama_stack/providers/inline/agents/meta_reference/openai_responses.py
@ -38,6 +38,7 @@ from llama_stack.apis.agents.openai_responses import (
    OpenAIResponseOutputMessageContent,
    OpenAIResponseOutputMessageContentOutputText,
    OpenAIResponseOutputMessageFileSearchToolCall,
+    OpenAIResponseOutputMessageFileSearchToolCallResults,
    OpenAIResponseOutputMessageFunctionToolCall,
    OpenAIResponseOutputMessageMCPListTools,
    OpenAIResponseOutputMessageWebSearchToolCall,
@ -333,6 +334,7 @@ class OpenAIResponsesImpl:
        temperature: float | None = None,
        text: OpenAIResponseText | None = None,
        tools: list[OpenAIResponseInputTool] | None = None,
+        include: list[str] | None = None,
        max_infer_iters: int | None = 10,
    ):
        stream = bool(stream)
@ -486,8 +488,12 @@ class OpenAIResponsesImpl:
            # Convert collected chunks to complete response
            if chat_response_tool_calls:
                tool_calls = [chat_response_tool_calls[i] for i in sorted(chat_response_tool_calls.keys())]
+
+                # when there are tool calls, we need to clear the content
+                chat_response_content = []
            else:
                tool_calls = None
+
            assistant_message = OpenAIAssistantMessageParam(
                content="".join(chat_response_content),
                tool_calls=tool_calls,
@ -826,12 +832,13 @@ class OpenAIResponsesImpl:
                        text = result.metadata["chunks"][i] if "chunks" in result.metadata else None
                        score = result.metadata["scores"][i] if "scores" in result.metadata else None
                        message.results.append(
-                            {
-                                "file_id": doc_id,
-                                "filename": doc_id,
-                                "text": text,
-                                "score": score,
-                            }
+                            OpenAIResponseOutputMessageFileSearchToolCallResults(
+                                file_id=doc_id,
+                                filename=doc_id,
+                                text=text,
+                                score=score,
+                                attributes={},
+                            )
                        )
                if error_exc or (result.error_code and result.error_code > 0) or result.error_message:
                    message.status = "failed"
--- a/llama_stack/providers/inline/safety/prompt_guard/prompt_guard.py
+++ b/llama_stack/providers/inline/safety/prompt_guard/prompt_guard.py
@ -15,6 +15,7 @@ from llama_stack.apis.safety import (
    RunShieldResponse,
    Safety,
    SafetyViolation,
+    ShieldStore,
    ViolationLevel,
 )
 from llama_stack.apis.shields import Shield
@ -32,6 +33,8 @@ PROMPT_GUARD_MODEL = "Prompt-Guard-86M"


 class PromptGuardSafetyImpl(Safety, ShieldsProtocolPrivate):
+    shield_store: ShieldStore
+
    def __init__(self, config: PromptGuardConfig, _deps) -> None:
        self.config = config

@ -53,7 +56,7 @@ class PromptGuardSafetyImpl(Safety, ShieldsProtocolPrivate):
        self,
        shield_id: str,
        messages: list[Message],
-        params: dict[str, Any] = None,
+        params: dict[str, Any],
    ) -> RunShieldResponse:
        shield = await self.shield_store.get_shield(shield_id)
        if not shield:
@ -61,6 +64,9 @@ class PromptGuardSafetyImpl(Safety, ShieldsProtocolPrivate):

        return await self.shield.run(messages)

+    async def run_moderation(self, input: str | list[str], model: str):
+        raise NotImplementedError("run_moderation not implemented for PromptGuard")
+

 class PromptGuardShield:
    def __init__(
@ -117,8 +123,10 @@ class PromptGuardShield:
        elif self.config.guard_type == PromptGuardType.jailbreak.value and score_malicious > self.threshold:
            violation = SafetyViolation(
                violation_level=ViolationLevel.ERROR,
-                violation_type=f"prompt_injection:malicious={score_malicious}",
-                violation_return_message="Sorry, I cannot do this.",
+                user_message="Sorry, I cannot do this.",
+                metadata={
+                    "violation_type": f"prompt_injection:malicious={score_malicious}",
+                },
            )

        return RunShieldResponse(violation=violation)
--- a/llama_stack/providers/inline/vector_io/faiss/faiss.py
+++ b/llama_stack/providers/inline/vector_io/faiss/faiss.py
@ -33,6 +33,7 @@ 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 (
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -128,11 +129,12 @@ class FaissIndex(EmbeddingIndex):
        # Save updated index
        await self._save_index()

-    async def delete_chunk(self, chunk_id: str) -> None:
-        if chunk_id not in self.chunk_ids:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        chunk_ids = [c.chunk_id for c in chunks_for_deletion]
+        if not set(chunk_ids).issubset(self.chunk_ids):
            return

-        async with self.chunk_id_lock:
+        def remove_chunk(chunk_id: str):
            index = self.chunk_ids.index(chunk_id)
            self.index.remove_ids(np.array([index]))

@ -146,6 +148,10 @@ class FaissIndex(EmbeddingIndex):
            self.chunk_by_index = new_chunk_by_index
            self.chunk_ids.pop(index)

+        async with self.chunk_id_lock:
+            for chunk_id in chunk_ids:
+                remove_chunk(chunk_id)
+
        await self._save_index()

    async def query_vector(
@ -174,7 +180,9 @@ class FaissIndex(EmbeddingIndex):
        k: int,
        score_threshold: float,
    ) -> QueryChunksResponse:
-        raise NotImplementedError("Keyword search is not supported in FAISS")
+        raise NotImplementedError(
+            "Keyword search is not supported - underlying DB FAISS does not support this search mode"
+        )

    async def query_hybrid(
        self,
@ -185,7 +193,9 @@ class FaissIndex(EmbeddingIndex):
        reranker_type: str,
        reranker_params: dict[str, Any] | None = None,
    ) -> QueryChunksResponse:
-        raise NotImplementedError("Hybrid search is not supported in FAISS")
+        raise NotImplementedError(
+            "Hybrid search is not supported - underlying DB FAISS does not support this search mode"
+        )


 class FaissVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolPrivate):
@ -293,8 +303,7 @@ class FaissVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolPr

        return await index.query_chunks(query, params)

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
-        """Delete a chunk from a faiss index"""
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        """Delete chunks from a faiss index"""
        faiss_index = self.cache[store_id].index
-        for chunk_id in chunk_ids:
-            await faiss_index.delete_chunk(chunk_id)
+        await faiss_index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/inline/vector_io/sqlite_vec/sqlite_vec.py
+++ b/llama_stack/providers/inline/vector_io/sqlite_vec/sqlite_vec.py
@ -31,6 +31,7 @@ from llama_stack.providers.utils.memory.openai_vector_store_mixin import OpenAIV
 from llama_stack.providers.utils.memory.vector_store import (
    RERANKER_TYPE_RRF,
    RERANKER_TYPE_WEIGHTED,
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -426,34 +427,36 @@ class SQLiteVecIndex(EmbeddingIndex):

        return QueryChunksResponse(chunks=chunks, scores=scores)

-    async def delete_chunk(self, chunk_id: str) -> None:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Remove a chunk from the SQLite vector store."""
+        chunk_ids = [c.chunk_id for c in chunks_for_deletion]

-        def _delete_chunk():
+        def _delete_chunks():
            connection = _create_sqlite_connection(self.db_path)
            cur = connection.cursor()
            try:
                cur.execute("BEGIN TRANSACTION")

                # Delete from metadata table
-                cur.execute(f"DELETE FROM {self.metadata_table} WHERE id = ?", (chunk_id,))
+                placeholders = ",".join("?" * len(chunk_ids))
+                cur.execute(f"DELETE FROM {self.metadata_table} WHERE id IN ({placeholders})", chunk_ids)

                # Delete from vector table
-                cur.execute(f"DELETE FROM {self.vector_table} WHERE id = ?", (chunk_id,))
+                cur.execute(f"DELETE FROM {self.vector_table} WHERE id IN ({placeholders})", chunk_ids)

                # Delete from FTS table
-                cur.execute(f"DELETE FROM {self.fts_table} WHERE id = ?", (chunk_id,))
+                cur.execute(f"DELETE FROM {self.fts_table} WHERE id IN ({placeholders})", chunk_ids)

                connection.commit()
            except Exception as e:
                connection.rollback()
-                logger.error(f"Error deleting chunk {chunk_id}: {e}")
+                logger.error(f"Error deleting chunks: {e}")
                raise
            finally:
                cur.close()
                connection.close()

-        await asyncio.to_thread(_delete_chunk)
+        await asyncio.to_thread(_delete_chunks)


 class SQLiteVecVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolPrivate):
@ -551,12 +554,10 @@ class SQLiteVecVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoc
            raise VectorStoreNotFoundError(vector_db_id)
        return await index.query_chunks(query, params)

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
-        """Delete a chunk from a sqlite_vec index."""
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        """Delete chunks from a sqlite_vec index."""
        index = await self._get_and_cache_vector_db_index(store_id)
        if not index:
            raise VectorStoreNotFoundError(store_id)

-        for chunk_id in chunk_ids:
-            # Use the index's delete_chunk method
-            await index.index.delete_chunk(chunk_id)
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/registry/inference.py
+++ b/llama_stack/providers/registry/inference.py
@ -213,6 +213,36 @@ def available_providers() -> list[ProviderSpec]:
                description="Google Gemini inference provider for accessing Gemini models and Google's AI services.",
            ),
        ),
+        remote_provider_spec(
+            api=Api.inference,
+            adapter=AdapterSpec(
+                adapter_type="vertexai",
+                pip_packages=["litellm", "google-cloud-aiplatform"],
+                module="llama_stack.providers.remote.inference.vertexai",
+                config_class="llama_stack.providers.remote.inference.vertexai.VertexAIConfig",
+                provider_data_validator="llama_stack.providers.remote.inference.vertexai.config.VertexAIProviderDataValidator",
+                description="""Google Vertex AI inference provider enables you to use Google's Gemini models through Google Cloud's Vertex AI platform, providing several advantages:
+
+• Enterprise-grade security: Uses Google Cloud's security controls and IAM
+• Better integration: Seamless integration with other Google Cloud services
+• Advanced features: Access to additional Vertex AI features like model tuning and monitoring
+• Authentication: Uses Google Cloud Application Default Credentials (ADC) instead of API keys
+
+Configuration:
+- Set VERTEX_AI_PROJECT environment variable (required)
+- Set VERTEX_AI_LOCATION environment variable (optional, defaults to us-central1)
+- Use Google Cloud Application Default Credentials or service account key
+
+Authentication Setup:
+Option 1 (Recommended): gcloud auth application-default login
+Option 2: Set GOOGLE_APPLICATION_CREDENTIALS to service account key path
+
+Available Models:
+- vertex_ai/gemini-2.0-flash
+- vertex_ai/gemini-2.5-flash
+- vertex_ai/gemini-2.5-pro""",
+            ),
+        ),
        remote_provider_spec(
            api=Api.inference,
            adapter=AdapterSpec(
--- a/llama_stack/providers/registry/vector_io.py
+++ b/llama_stack/providers/registry/vector_io.py
@ -45,6 +45,18 @@ That means you'll get fast and efficient vector retrieval.
 - Lightweight and easy to use
 - Fully integrated with Llama Stack
 - GPU support
+- **Vector search** - FAISS supports pure vector similarity search using embeddings
+
+## Search Modes
+
+**Supported:**
+- **Vector Search** (`mode="vector"`): Performs vector similarity search using embeddings
+
+**Not Supported:**
+- **Keyword Search** (`mode="keyword"`): Not supported by FAISS
+- **Hybrid Search** (`mode="hybrid"`): Not supported by FAISS
+
+> **Note**: FAISS is designed as a pure vector similarity search library. See the [FAISS GitHub repository](https://github.com/facebookresearch/faiss) for more details about FAISS's core functionality.

 ## Usage

@ -330,6 +342,7 @@ See [Chroma's documentation](https://docs.trychroma.com/docs/overview/introducti
 """,
            ),
            api_dependencies=[Api.inference],
+            optional_api_dependencies=[Api.files],
        ),
        InlineProviderSpec(
            api=Api.vector_io,
@ -338,6 +351,7 @@ See [Chroma's documentation](https://docs.trychroma.com/docs/overview/introducti
            module="llama_stack.providers.inline.vector_io.chroma",
            config_class="llama_stack.providers.inline.vector_io.chroma.ChromaVectorIOConfig",
            api_dependencies=[Api.inference],
+            optional_api_dependencies=[Api.files],
            description="""
 [Chroma](https://www.trychroma.com/) is an inline and remote vector
 database provider for Llama Stack. It allows you to store and query vectors directly within a Chroma database.
@ -452,6 +466,7 @@ See [Weaviate's documentation](https://weaviate.io/developers/weaviate) for more
 """,
            ),
            api_dependencies=[Api.inference],
+            optional_api_dependencies=[Api.files],
        ),
        InlineProviderSpec(
            api=Api.vector_io,
@ -535,6 +550,7 @@ That means you're not limited to storing vectors in memory or in a separate serv

 - Easy to use
 - Fully integrated with Llama Stack
+- Supports all search modes: vector, keyword, and hybrid search (both inline and remote configurations)

 ## Usage

@ -625,6 +641,92 @@ vector_io:
 - **`client_pem_path`**: Path to the **client certificate** file (required for mTLS).
 - **`client_key_path`**: Path to the **client private key** file (required for mTLS).

+## Search Modes
+
+Milvus supports three different search modes for both inline and remote configurations:
+
+### Vector Search
+Vector search uses semantic similarity to find the most relevant chunks based on embedding vectors. This is the default search mode and works well for finding conceptually similar content.
+
+```python
+# Vector search example
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="What is machine learning?",
+    search_mode="vector",
+    max_num_results=5,
+)
+```
+
+### Keyword Search
+Keyword search uses traditional text-based matching to find chunks containing specific terms or phrases. This is useful when you need exact term matches.
+
+```python
+# Keyword search example
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="Python programming language",
+    search_mode="keyword",
+    max_num_results=5,
+)
+```
+
+### Hybrid Search
+Hybrid search combines both vector and keyword search methods to provide more comprehensive results. It leverages the strengths of both semantic similarity and exact term matching.
+
+#### Basic Hybrid Search
+```python
+# Basic hybrid search example (uses RRF ranker with default impact_factor=60.0)
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+)
+```
+
+**Note**: The default `impact_factor` value of 60.0 was empirically determined to be optimal in the original RRF research paper: ["Reciprocal Rank Fusion outperforms Condorcet and individual Rank Learning Methods"](https://plg.uwaterloo.ca/~gvcormac/cormacksigir09-rrf.pdf) (Cormack et al., 2009).
+
+#### Hybrid Search with RRF (Reciprocal Rank Fusion) Ranker
+RRF combines rankings from vector and keyword search by using reciprocal ranks. The impact factor controls how much weight is given to higher-ranked results.
+
+```python
+# Hybrid search with custom RRF parameters
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+    ranking_options={
+        "ranker": {
+            "type": "rrf",
+            "impact_factor": 100.0,  # Higher values give more weight to top-ranked results
+        }
+    },
+)
+```
+
+#### Hybrid Search with Weighted Ranker
+Weighted ranker linearly combines normalized scores from vector and keyword search. The alpha parameter controls the balance between the two search methods.
+
+```python
+# Hybrid search with weighted ranker
+search_response = client.vector_stores.search(
+    vector_store_id=vector_store.id,
+    query="neural networks in Python",
+    search_mode="hybrid",
+    max_num_results=5,
+    ranking_options={
+        "ranker": {
+            "type": "weighted",
+            "alpha": 0.7,  # 70% vector search, 30% keyword search
+        }
+    },
+)
+```
+
+For detailed documentation on RRF and Weighted rankers, please refer to the [Milvus Reranking Guide](https://milvus.io/docs/reranking.md).
+
 ## Documentation
 See the [Milvus documentation](https://milvus.io/docs/install-overview.md) for more details about Milvus in general.

@ -632,6 +734,7 @@ For more details on TLS configuration, refer to the [TLS setup guide](https://mi
 """,
            ),
            api_dependencies=[Api.inference],
+            optional_api_dependencies=[Api.files],
        ),
        InlineProviderSpec(
            api=Api.vector_io,
--- a/llama_stack/providers/remote/inference/fireworks/fireworks.py
+++ b/llama_stack/providers/remote/inference/fireworks/fireworks.py
@ -235,6 +235,7 @@ class FireworksInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProv

        llama_model = self.get_llama_model(request.model)
        if isinstance(request, ChatCompletionRequest):
+            # TODO: tools are never added to the request, so we need to add them here
            if media_present or not llama_model:
                input_dict["messages"] = [
                    await convert_message_to_openai_dict(m, download=True) for m in request.messages
@ -378,6 +379,7 @@ class FireworksInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProv
        # Fireworks chat completions OpenAI-compatible API does not support
        # tool calls properly.
        llama_model = self.get_llama_model(model_obj.provider_resource_id)
+
        if llama_model:
            return await OpenAIChatCompletionToLlamaStackMixin.openai_chat_completion(
                self,
@ -431,4 +433,5 @@ class FireworksInferenceAdapter(ModelRegistryHelper, Inference, NeedsRequestProv
            user=user,
        )

+        logger.debug(f"fireworks params: {params}")
        return await self._get_openai_client().chat.completions.create(model=model_obj.provider_resource_id, **params)
--- a/llama_stack/providers/remote/inference/ollama/ollama.py
+++ b/llama_stack/providers/remote/inference/ollama/ollama.py
@ -457,9 +457,6 @@ class OllamaInferenceAdapter(
        user: str | None = None,
    ) -> OpenAIEmbeddingsResponse:
        model_obj = await self._get_model(model)
-        if model_obj.model_type != ModelType.embedding:
-            raise ValueError(f"Model {model} is not an embedding model")
-
        if model_obj.provider_resource_id is None:
            raise ValueError(f"Model {model} has no provider_resource_id set")

--- a/llama_stack/providers/remote/inference/vertexai/init.py
+++ b/llama_stack/providers/remote/inference/vertexai/init.py
@ -0,0 +1,15 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .config import VertexAIConfig
+
+
+async def get_adapter_impl(config: VertexAIConfig, _deps):
+    from .vertexai import VertexAIInferenceAdapter
+
+    impl = VertexAIInferenceAdapter(config)
+    await impl.initialize()
+    return impl
--- a/llama_stack/providers/remote/inference/vertexai/config.py
+++ b/llama_stack/providers/remote/inference/vertexai/config.py
@ -0,0 +1,45 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from typing import Any
+
+from pydantic import BaseModel, Field
+
+from llama_stack.schema_utils import json_schema_type
+
+
+class VertexAIProviderDataValidator(BaseModel):
+    vertex_project: str | None = Field(
+        default=None,
+        description="Google Cloud project ID for Vertex AI",
+    )
+    vertex_location: str | None = Field(
+        default=None,
+        description="Google Cloud location for Vertex AI (e.g., us-central1)",
+    )
+
+
+@json_schema_type
+class VertexAIConfig(BaseModel):
+    project: str = Field(
+        description="Google Cloud project ID for Vertex AI",
+    )
+    location: str = Field(
+        default="us-central1",
+        description="Google Cloud location for Vertex AI",
+    )
+
+    @classmethod
+    def sample_run_config(
+        cls,
+        project: str = "${env.VERTEX_AI_PROJECT:=}",
+        location: str = "${env.VERTEX_AI_LOCATION:=us-central1}",
+        **kwargs,
+    ) -> dict[str, Any]:
+        return {
+            "project": project,
+            "location": location,
+        }
--- a/llama_stack/providers/remote/inference/vertexai/models.py
+++ b/llama_stack/providers/remote/inference/vertexai/models.py
@ -0,0 +1,20 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from llama_stack.providers.utils.inference.model_registry import (
+    ProviderModelEntry,
+)
+
+# Vertex AI model IDs with vertex_ai/ prefix as required by litellm
+LLM_MODEL_IDS = [
+    "vertex_ai/gemini-2.0-flash",
+    "vertex_ai/gemini-2.5-flash",
+    "vertex_ai/gemini-2.5-pro",
+]
+
+SAFETY_MODELS_ENTRIES = list[ProviderModelEntry]()
+
+MODEL_ENTRIES = [ProviderModelEntry(provider_model_id=m) for m in LLM_MODEL_IDS] + SAFETY_MODELS_ENTRIES
--- a/llama_stack/providers/remote/inference/vertexai/vertexai.py
+++ b/llama_stack/providers/remote/inference/vertexai/vertexai.py
@ -0,0 +1,52 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from typing import Any
+
+from llama_stack.apis.inference import ChatCompletionRequest
+from llama_stack.providers.utils.inference.litellm_openai_mixin import (
+    LiteLLMOpenAIMixin,
+)
+
+from .config import VertexAIConfig
+from .models import MODEL_ENTRIES
+
+
+class VertexAIInferenceAdapter(LiteLLMOpenAIMixin):
+    def __init__(self, config: VertexAIConfig) -> None:
+        LiteLLMOpenAIMixin.__init__(
+            self,
+            MODEL_ENTRIES,
+            litellm_provider_name="vertex_ai",
+            api_key_from_config=None,  # Vertex AI uses ADC, not API keys
+            provider_data_api_key_field="vertex_project",  # Use project for validation
+        )
+        self.config = config
+
+    def get_api_key(self) -> str:
+        # Vertex AI doesn't use API keys, it uses Application Default Credentials
+        # Return empty string to let litellm handle authentication via ADC
+        return ""
+
+    async def _get_params(self, request: ChatCompletionRequest) -> dict[str, Any]:
+        # Get base parameters from parent
+        params = await super()._get_params(request)
+
+        # Add Vertex AI specific parameters
+        provider_data = self.get_request_provider_data()
+        if provider_data:
+            if getattr(provider_data, "vertex_project", None):
+                params["vertex_project"] = provider_data.vertex_project
+            if getattr(provider_data, "vertex_location", None):
+                params["vertex_location"] = provider_data.vertex_location
+        else:
+            params["vertex_project"] = self.config.project
+            params["vertex_location"] = self.config.location
+
+        # Remove api_key since Vertex AI uses ADC
+        params.pop("api_key", None)
+
+        return params
--- a/llama_stack/providers/remote/vector_io/chroma/chroma.py
+++ b/llama_stack/providers/remote/vector_io/chroma/chroma.py
@ -26,6 +26,7 @@ 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 (
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -115,8 +116,10 @@ class ChromaIndex(EmbeddingIndex):
    ) -> QueryChunksResponse:
        raise NotImplementedError("Keyword search is not supported in Chroma")

-    async def delete_chunk(self, chunk_id: str) -> None:
-        raise NotImplementedError("delete_chunk is not supported in Chroma")
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        """Delete a single chunk from the Chroma collection by its ID."""
+        ids = [f"{chunk.document_id}:{chunk.chunk_id}" for chunk in chunks_for_deletion]
+        await maybe_await(self.collection.delete(ids=ids))

    async def query_hybrid(
        self,
@ -144,6 +147,7 @@ class ChromaVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
        self.cache = {}
        self.kvstore: KVStore | None = None
        self.vector_db_store = None
+        self.files_api = files_api

    async def initialize(self) -> None:
        self.kvstore = await kvstore_impl(self.config.kvstore)
@ -227,5 +231,10 @@ class ChromaVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
        self.cache[vector_db_id] = index
        return index

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
-        raise NotImplementedError("OpenAI Vector Stores API is not supported in Chroma")
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        """Delete chunks from a Chroma vector store."""
+        index = await self._get_and_cache_vector_db_index(store_id)
+        if not index:
+            raise ValueError(f"Vector DB {store_id} not found")
+
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/remote/vector_io/milvus/milvus.py
+++ b/llama_stack/providers/remote/vector_io/milvus/milvus.py
@ -28,6 +28,7 @@ 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 (
    RERANKER_TYPE_WEIGHTED,
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -287,14 +288,17 @@ class MilvusIndex(EmbeddingIndex):

        return QueryChunksResponse(chunks=filtered_chunks, scores=filtered_scores)

-    async def delete_chunk(self, chunk_id: str) -> None:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Remove a chunk from the Milvus collection."""
+        chunk_ids = [c.chunk_id for c in chunks_for_deletion]
        try:
+            # Use IN clause with square brackets and single quotes for VARCHAR field
+            chunk_ids_str = ", ".join(f"'{chunk_id}'" for chunk_id in chunk_ids)
            await asyncio.to_thread(
-                self.client.delete, collection_name=self.collection_name, filter=f'chunk_id == "{chunk_id}"'
+                self.client.delete, collection_name=self.collection_name, filter=f"chunk_id in [{chunk_ids_str}]"
            )
        except Exception as e:
-            logger.error(f"Error deleting chunk {chunk_id} from Milvus collection {self.collection_name}: {e}")
+            logger.error(f"Error deleting chunks from Milvus collection {self.collection_name}: {e}")
            raise


@ -420,12 +424,10 @@ class MilvusVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP

        return await index.query_chunks(query, params)

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Delete a chunk from a milvus vector store."""
        index = await self._get_and_cache_vector_db_index(store_id)
        if not index:
            raise VectorStoreNotFoundError(store_id)

-        for chunk_id in chunk_ids:
-            # Use the index's delete_chunk method
-            await index.index.delete_chunk(chunk_id)
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/remote/vector_io/pgvector/pgvector.py
+++ b/llama_stack/providers/remote/vector_io/pgvector/pgvector.py
@ -27,6 +27,7 @@ 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 (
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -163,10 +164,11 @@ class PGVectorIndex(EmbeddingIndex):
        with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
            cur.execute(f"DROP TABLE IF EXISTS {self.table_name}")

-    async def delete_chunk(self, chunk_id: str) -> None:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Remove a chunk from the PostgreSQL table."""
+        chunk_ids = [c.chunk_id for c in chunks_for_deletion]
        with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
-            cur.execute(f"DELETE FROM {self.table_name} WHERE id = %s", (chunk_id,))
+            cur.execute(f"DELETE FROM {self.table_name} WHERE id = ANY(%s)", (chunk_ids,))


 class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolPrivate):
@ -275,12 +277,10 @@ class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoco
        self.cache[vector_db_id] = VectorDBWithIndex(vector_db, index, self.inference_api)
        return self.cache[vector_db_id]

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Delete a chunk from a PostgreSQL vector store."""
        index = await self._get_and_cache_vector_db_index(store_id)
        if not index:
            raise VectorStoreNotFoundError(store_id)

-        for chunk_id in chunk_ids:
-            # Use the index's delete_chunk method
-            await index.index.delete_chunk(chunk_id)
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/remote/vector_io/qdrant/qdrant.py
+++ b/llama_stack/providers/remote/vector_io/qdrant/qdrant.py
@ -29,6 +29,7 @@ from llama_stack.providers.inline.vector_io.qdrant import QdrantVectorIOConfig a
 from llama_stack.providers.utils.kvstore import KVStore, kvstore_impl
 from llama_stack.providers.utils.memory.openai_vector_store_mixin import OpenAIVectorStoreMixin
 from llama_stack.providers.utils.memory.vector_store import (
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -88,15 +89,16 @@ class QdrantIndex(EmbeddingIndex):

        await self.client.upsert(collection_name=self.collection_name, points=points)

-    async def delete_chunk(self, chunk_id: str) -> None:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Remove a chunk from the Qdrant collection."""
+        chunk_ids = [convert_id(c.chunk_id) for c in chunks_for_deletion]
        try:
            await self.client.delete(
                collection_name=self.collection_name,
-                points_selector=models.PointIdsList(points=[convert_id(chunk_id)]),
+                points_selector=models.PointIdsList(points=chunk_ids),
            )
        except Exception as e:
-            log.error(f"Error deleting chunk {chunk_id} from Qdrant collection {self.collection_name}: {e}")
+            log.error(f"Error deleting chunks from Qdrant collection {self.collection_name}: {e}")
            raise

    async def query_vector(self, embedding: NDArray, k: int, score_threshold: float) -> QueryChunksResponse:
@ -264,12 +266,14 @@ class QdrantVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolP
    ) -> VectorStoreFileObject:
        # Qdrant doesn't allow multiple clients to access the same storage path simultaneously.
        async with self._qdrant_lock:
-            await super().openai_attach_file_to_vector_store(vector_store_id, file_id, attributes, chunking_strategy)
+            return await super().openai_attach_file_to_vector_store(
+                vector_store_id, file_id, attributes, chunking_strategy
+            )

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        """Delete chunks from a Qdrant vector store."""
        index = await self._get_and_cache_vector_db_index(store_id)
        if not index:
            raise ValueError(f"Vector DB {store_id} not found")
-        for chunk_id in chunk_ids:
-            await index.index.delete_chunk(chunk_id)
+
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/remote/vector_io/weaviate/weaviate.py
+++ b/llama_stack/providers/remote/vector_io/weaviate/weaviate.py
@ -26,6 +26,7 @@ from llama_stack.providers.utils.memory.openai_vector_store_mixin import (
    OpenAIVectorStoreMixin,
 )
 from llama_stack.providers.utils.memory.vector_store import (
+    ChunkForDeletion,
    EmbeddingIndex,
    VectorDBWithIndex,
 )
@ -67,6 +68,7 @@ class WeaviateIndex(EmbeddingIndex):
            data_objects.append(
                wvc.data.DataObject(
                    properties={
+                        "chunk_id": chunk.chunk_id,
                        "chunk_content": chunk.model_dump_json(),
                    },
                    vector=embeddings[i].tolist(),
@ -79,10 +81,11 @@ class WeaviateIndex(EmbeddingIndex):
        # TODO: make this async friendly
        collection.data.insert_many(data_objects)

-    async def delete_chunk(self, chunk_id: str) -> None:
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        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]))
+        chunk_ids = [chunk.chunk_id for chunk in chunks_for_deletion]
+        collection.data.delete_many(where=Filter.by_property("chunk_id").contains_any(chunk_ids))

    async def query_vector(self, embedding: NDArray, k: int, score_threshold: float) -> QueryChunksResponse:
        sanitized_collection_name = sanitize_collection_name(self.collection_name, weaviate_format=True)
@ -307,10 +310,10 @@ class WeaviateVectorIOAdapter(

        return await index.query_chunks(query, params)

-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
        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)
+        await index.index.delete_chunks(chunks_for_deletion)
--- a/llama_stack/providers/utils/inference/openai_compat.py
+++ b/llama_stack/providers/utils/inference/openai_compat.py
@ -70,7 +70,7 @@ from openai.types.chat.chat_completion_chunk import (
 from openai.types.chat.chat_completion_content_part_image_param import (
    ImageURL as OpenAIImageURL,
 )
-from openai.types.chat.chat_completion_message_tool_call_param import (
+from openai.types.chat.chat_completion_message_tool_call import (
    Function as OpenAIFunction,
 )
 from pydantic import BaseModel
--- a/llama_stack/providers/utils/memory/openai_vector_store_mixin.py
+++ b/llama_stack/providers/utils/memory/openai_vector_store_mixin.py
@ -6,7 +6,6 @@

 import asyncio
 import json
-import logging
 import mimetypes
 import time
 import uuid
@ -37,10 +36,15 @@ from llama_stack.apis.vector_io import (
    VectorStoreSearchResponse,
    VectorStoreSearchResponsePage,
 )
+from llama_stack.log import get_logger
 from llama_stack.providers.utils.kvstore.api import KVStore
-from llama_stack.providers.utils.memory.vector_store import content_from_data_and_mime_type, make_overlapped_chunks
+from llama_stack.providers.utils.memory.vector_store import (
+    ChunkForDeletion,
+    content_from_data_and_mime_type,
+    make_overlapped_chunks,
+)

-logger = logging.getLogger(__name__)
+logger = get_logger(__name__, category="vector_io")

 # Constants for OpenAI vector stores
 CHUNK_MULTIPLIER = 5
@ -154,8 +158,8 @@ class OpenAIVectorStoreMixin(ABC):
        self.openai_vector_stores = await self._load_openai_vector_stores()

    @abstractmethod
-    async def delete_chunks(self, store_id: str, chunk_ids: list[str]) -> None:
-        """Delete a chunk from a vector store."""
+    async def delete_chunks(self, store_id: str, chunks_for_deletion: list[ChunkForDeletion]) -> None:
+        """Delete chunks from a vector store."""
        pass

    @abstractmethod
@ -614,7 +618,7 @@ class OpenAIVectorStoreMixin(ABC):
                )
                vector_store_file_object.status = "completed"
        except Exception as e:
-            logger.error(f"Error attaching file to vector store: {e}")
+            logger.exception("Error attaching file to vector store")
            vector_store_file_object.status = "failed"
            vector_store_file_object.last_error = VectorStoreFileLastError(
                code="server_error",
@ -767,7 +771,21 @@ class OpenAIVectorStoreMixin(ABC):

        dict_chunks = await self._load_openai_vector_store_file_contents(vector_store_id, file_id)
        chunks = [Chunk.model_validate(c) for c in dict_chunks]
-        await self.delete_chunks(vector_store_id, [str(c.chunk_id) for c in chunks if c.chunk_id])
+
+        # Create ChunkForDeletion objects with both chunk_id and document_id
+        chunks_for_deletion = []
+        for c in chunks:
+            if c.chunk_id:
+                document_id = c.metadata.get("document_id") or (
+                    c.chunk_metadata.document_id if c.chunk_metadata else None
+                )
+                if document_id:
+                    chunks_for_deletion.append(ChunkForDeletion(chunk_id=str(c.chunk_id), document_id=document_id))
+                else:
+                    logger.warning(f"Chunk {c.chunk_id} has no document_id, skipping deletion")
+
+        if chunks_for_deletion:
+            await self.delete_chunks(vector_store_id, chunks_for_deletion)

        store_info = self.openai_vector_stores[vector_store_id].copy()

--- a/llama_stack/providers/utils/memory/vector_store.py
+++ b/llama_stack/providers/utils/memory/vector_store.py
@ -16,6 +16,7 @@ from urllib.parse import unquote
 import httpx
 import numpy as np
 from numpy.typing import NDArray
+from pydantic import BaseModel

 from llama_stack.apis.common.content_types import (
    URL,
@ -34,6 +35,18 @@ from llama_stack.providers.utils.vector_io.vector_utils import generate_chunk_id

 log = logging.getLogger(__name__)

+
+class ChunkForDeletion(BaseModel):
+    """Information needed to delete a chunk from a vector store.
+
+    :param chunk_id: The ID of the chunk to delete
+    :param document_id: The ID of the document this chunk belongs to
+    """
+
+    chunk_id: str
+    document_id: str
+
+
 # Constants for reranker types
 RERANKER_TYPE_RRF = "rrf"
 RERANKER_TYPE_WEIGHTED = "weighted"
@ -232,7 +245,7 @@ class EmbeddingIndex(ABC):
        raise NotImplementedError()

    @abstractmethod
-    async def delete_chunk(self, chunk_id: str):
+    async def delete_chunks(self, chunks_for_deletion: list[ChunkForDeletion]):
        raise NotImplementedError()

    @abstractmethod
--- a/llama_stack/ui/app/chat-playground/page.tsx
+++ b/llama_stack/ui/app/chat-playground/page.tsx
@ -175,7 +175,7 @@ const handleSubmitWithContent = async (content: string) => {
  return (
    <div className="flex flex-col h-full max-w-4xl mx-auto">
      <div className="mb-4 flex justify-between items-center">
-        <h1 className="text-2xl font-bold">Chat Playground</h1>
+        <h1 className="text-2xl font-bold">Chat Playground (Completions)</h1>
        <div className="flex gap-2">
          <Select value={selectedModel} onValueChange={setSelectedModel} disabled={isModelsLoading || isGenerating}>
            <SelectTrigger className="w-[180px]">
--- a/llama_stack/ui/components/layout/app-sidebar.tsx
+++ b/llama_stack/ui/components/layout/app-sidebar.tsx
@ -6,6 +6,8 @@ import {
  MoveUpRight,
  Database,
  MessageCircle,
+  Settings2,
+  Compass,
 } from "lucide-react";
 import Link from "next/link";
 import { usePathname } from "next/navigation";
@ -22,15 +24,16 @@ import {
  SidebarMenuItem,
  SidebarHeader,
 } from "@/components/ui/sidebar";
-// Extracted Chat Playground item
-const chatPlaygroundItem = {
-  title: "Chat Playground",
-  url: "/chat-playground",
-  icon: MessageCircle,
-};

-// Removed Chat Playground from log items
-const logItems = [
+const createItems = [
+  {
+    title: "Chat Playground",
+    url: "/chat-playground",
+    icon: MessageCircle,
+  },
+];
+
+const manageItems = [
  {
    title: "Chat Completions",
    url: "/logs/chat-completions",
@ -53,77 +56,96 @@ const logItems = [
  },
 ];

+const optimizeItems: { title: string; url: string; icon: React.ElementType }[] = [
+    {
+        title: "Evaluations",
+        url: "",
+        icon: Compass,
+    },
+    {
+        title: "Fine-tuning",
+        url: "",
+        icon: Settings2,
+    },
+];
+
+interface SidebarItem {
+  title: string;
+  url: string;
+  icon: React.ElementType;
+}
+
 export function AppSidebar() {
  const pathname = usePathname();

-  return (
-    <Sidebar>
-      <SidebarHeader>
-        <Link href="/">Llama Stack</Link>
-      </SidebarHeader>
-      <SidebarContent>
-        {/* Chat Playground as its own section */}
-        <SidebarGroup>
-          <SidebarGroupContent>
-            <SidebarMenu>
-              <SidebarMenuItem>
+  const renderSidebarItems = (items: SidebarItem[]) => {
+    return items.map((item) => {
+      const isActive = pathname.startsWith(item.url);
+      return (
+        <SidebarMenuItem key={item.title}>
+          <SidebarMenuButton
+            asChild
+            className={cn(
+              "justify-start",
+              isActive &&
+                "bg-gray-200 dark:bg-gray-700 hover:bg-gray-200 dark:hover:bg-gray-700 text-gray-900 dark:text-gray-100",
+            )}
+          >
+            <Link href={item.url}>
+              <item.icon
+                className={cn(
+                  isActive && "text-gray-900 dark:text-gray-100",
+                  "mr-2 h-4 w-4",
+                )}
+              />
+              <span>{item.title}</span>
+            </Link>
+          </SidebarMenuButton>
+        </SidebarMenuItem>
+      );
+    });
+  };
+
+return (
+  <Sidebar>
+    <SidebarHeader>
+      <Link href="/">Llama Stack</Link>
+    </SidebarHeader>
+    <SidebarContent>
+      <SidebarGroup>
+        <SidebarGroupLabel>Create</SidebarGroupLabel>
+        <SidebarGroupContent>
+          <SidebarMenu>{renderSidebarItems(createItems)}</SidebarMenu>
+        </SidebarGroupContent>
+      </SidebarGroup>
+
+      <SidebarGroup>
+        <SidebarGroupLabel>Manage</SidebarGroupLabel>
+        <SidebarGroupContent>
+          <SidebarMenu>{renderSidebarItems(manageItems)}</SidebarMenu>
+        </SidebarGroupContent>
+      </SidebarGroup>
+
+      <SidebarGroup>
+        <SidebarGroupLabel>Optimize</SidebarGroupLabel>
+        <SidebarGroupContent>
+          <SidebarMenu>
+            {optimizeItems.map((item) => (
+              <SidebarMenuItem key={item.title}>
                <SidebarMenuButton
-                  asChild
-                  className={cn(
-                    "justify-start",
-                    pathname.startsWith(chatPlaygroundItem.url) &&
-                      "bg-gray-200 dark:bg-gray-700 hover:bg-gray-200 dark:hover:bg-gray-700 text-gray-900 dark:text-gray-100",
-                  )}
+                  disabled
+                  className="justify-start opacity-60 cursor-not-allowed"
                >
-                  <Link href={chatPlaygroundItem.url}>
-                    <chatPlaygroundItem.icon
-                      className={cn(
-                        pathname.startsWith(chatPlaygroundItem.url) && "text-gray-900 dark:text-gray-100",
-                        "mr-2 h-4 w-4",
-                      )}
-                    />
-                    <span>{chatPlaygroundItem.title}</span>
-                  </Link>
+                  <item.icon className="mr-2 h-4 w-4" />
+                  <span>{item.title}</span>
+                  <span className="ml-2 text-xs text-gray-500">(Coming Soon)</span>
                </SidebarMenuButton>
              </SidebarMenuItem>
-            </SidebarMenu>
-          </SidebarGroupContent>
-        </SidebarGroup>
-
-        {/* Logs section */}
-        <SidebarGroup>
-          <SidebarGroupLabel>Logs</SidebarGroupLabel>
-          <SidebarGroupContent>
-            <SidebarMenu>
-              {logItems.map((item) => {
-                const isActive = pathname.startsWith(item.url);
-                return (
-                  <SidebarMenuItem key={item.title}>
-                    <SidebarMenuButton
-                      asChild
-                      className={cn(
-                        "justify-start",
-                        isActive &&
-                          "bg-gray-200 dark:bg-gray-700 hover:bg-gray-200 dark:hover:bg-gray-700 text-gray-900 dark:text-gray-100",
-                      )}
-                    >
-                      <Link href={item.url}>
-                        <item.icon
-                          className={cn(
-                            isActive && "text-gray-900 dark:text-gray-100",
-                            "mr-2 h-4 w-4",
-                          )}
-                        />
-                        <span>{item.title}</span>
-                      </Link>
-                    </SidebarMenuButton>
-                  </SidebarMenuItem>
-                );
-              })}
-            </SidebarMenu>
-          </SidebarGroupContent>
-        </SidebarGroup>
-      </SidebarContent>
-    </Sidebar>
+            ))}
+          </SidebarMenu>
+        </SidebarGroupContent>
+      </SidebarGroup>
+    </SidebarContent>
+  </Sidebar>
  );
 }
--- a/pyproject.toml
+++ b/pyproject.toml
@ -33,7 +33,7 @@ dependencies = [
    "jsonschema",
    "llama-stack-client>=0.2.17",
    "llama-api-client>=0.1.2",
-    "openai>=1.66",
+    "openai>=1.99.6",
    "prompt-toolkit",
    "python-dotenv",
    "python-jose[cryptography]",
@ -266,7 +266,6 @@ exclude = [
    "^llama_stack/providers/inline/post_training/common/validator\\.py$",
    "^llama_stack/providers/inline/safety/code_scanner/",
    "^llama_stack/providers/inline/safety/llama_guard/",
-    "^llama_stack/providers/inline/safety/prompt_guard/",
    "^llama_stack/providers/inline/scoring/basic/",
    "^llama_stack/providers/inline/scoring/braintrust/",
    "^llama_stack/providers/inline/scoring/llm_as_judge/",
--- a/tests/common/mcp.py
+++ b/tests/common/mcp.py
@ -16,13 +16,10 @@ MCP_TOOLGROUP_ID = "mcp::localmcp"

 def default_tools():
    """Default tools for backward compatibility."""
-    from mcp import types
    from mcp.server.fastmcp import Context

-    async def greet_everyone(
-        url: str, ctx: Context
-    ) -> list[types.TextContent | types.ImageContent | types.EmbeddedResource]:
-        return [types.TextContent(type="text", text="Hello, world!")]
+    async def greet_everyone(url: str, ctx: Context) -> str:
+        return "Hello, world!"

    async def get_boiling_point(liquid_name: str, celsius: bool = True) -> int:
        """
@ -45,7 +42,6 @@ def default_tools():

 def dependency_tools():
    """Tools with natural dependencies for multi-turn testing."""
-    from mcp import types
    from mcp.server.fastmcp import Context

    async def get_user_id(username: str, ctx: Context) -> str:
@ -106,7 +102,7 @@ def dependency_tools():
        else:
            access = "no"

-        return [types.TextContent(type="text", text=access)]
+        return access

    async def get_experiment_id(experiment_name: str, ctx: Context) -> str:
        """
@ -245,7 +241,6 @@ def make_mcp_server(required_auth_token: str | None = None, tools: dict[str, Cal
    try:
        yield {"server_url": server_url}
    finally:
-        print("Telling SSE server to exit")
        server_instance.should_exit = True
        time.sleep(0.5)

@ -269,4 +264,3 @@ def make_mcp_server(required_auth_token: str | None = None, tools: dict[str, Cal

        AppStatus.should_exit = False
        AppStatus.should_exit_event = None
-        print("SSE server exited")
--- a/tests/external/llama-stack-api-weather/pyproject.toml
+++ b/tests/external/llama-stack-api-weather/pyproject.toml
@ -3,7 +3,7 @@ name = "llama-stack-api-weather"
 version = "0.1.0"
 description = "Weather API for Llama Stack"
 readme = "README.md"
-requires-python = ">=3.10"
+requires-python = ">=3.12"
 dependencies = ["llama-stack", "pydantic"]

 [build-system]
--- a/tests/external/llama-stack-provider-kaze/pyproject.toml
+++ b/tests/external/llama-stack-provider-kaze/pyproject.toml
@ -3,7 +3,7 @@ name = "llama-stack-provider-kaze"
 version = "0.1.0"
 description = "Kaze weather provider for Llama Stack"
 readme = "README.md"
-requires-python = ">=3.10"
+requires-python = ">=3.12"
 dependencies = ["llama-stack", "pydantic", "aiohttp"]

 [build-system]
--- a/tests/integration/fixtures/common.py
+++ b/tests/integration/fixtures/common.py
@ -270,7 +270,7 @@ def openai_client(client_with_models):

@pytest.fixture(params=["openai_client", "client_with_models"])
 def compat_client(request, client_with_models):
-    if isinstance(client_with_models, LlamaStackAsLibraryClient):
+    if request.param == "openai_client" and isinstance(client_with_models, LlamaStackAsLibraryClient):
        # OpenAI client expects a server, so unless we also rewrite OpenAI client's requests
        # to go via the Stack library client (which itself rewrites requests to be served inline),
        # we cannot do this.
--- a/tests/integration/inference/test_openai_completion.py
+++ b/tests/integration/inference/test_openai_completion.py
@ -34,6 +34,7 @@ def skip_if_model_doesnt_support_openai_completion(client_with_models, model_id)
        "remote::runpod",
        "remote::sambanova",
        "remote::tgi",
+        "remote::vertexai",
    ):
        pytest.skip(f"Model {model_id} hosted by {provider.provider_type} doesn't support OpenAI completions.")

--- a/tests/integration/inference/test_text_inference.py
+++ b/tests/integration/inference/test_text_inference.py
@ -29,6 +29,7 @@ def skip_if_model_doesnt_support_completion(client_with_models, model_id):
            "remote::openai",
            "remote::anthropic",
            "remote::gemini",
+            "remote::vertexai",
            "remote::groq",
            "remote::sambanova",
        )
--- a/tests/integration/non_ci/responses/fixtures/test_cases/responses.yaml
+++ b/tests/integration/non_ci/responses/fixtures/test_cases/responses.yaml
@ -137,7 +137,7 @@ test_response_multi_turn_tool_execution:
        server_url: "<FILLED_BY_TEST_RUNNER>"
      output: "yes"
    - case_id: "experiment_results_lookup"
-      input: "I need to get the results for the 'boiling_point' experiment. First, get the experiment ID for 'boiling_point', then use that ID to get the experiment results. Tell me what you found."
+      input: "I need to get the results for the 'boiling_point' experiment. First, get the experiment ID for 'boiling_point', then use that ID to get the experiment results. Tell me the boiling point in Celsius."
      tools:
      - type: mcp
        server_label: "localmcp"
@ -149,7 +149,7 @@ test_response_multi_turn_tool_execution_streaming:
  test_params:
    case:
    - case_id: "user_permissions_workflow"
-      input: "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step."
+      input: "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
      tools:
      - type: mcp
        server_label: "localmcp"
@ -157,7 +157,7 @@ test_response_multi_turn_tool_execution_streaming:
      stream: true
      output: "no"
    - case_id: "experiment_analysis_streaming"
-      input: "I need a complete analysis: First, get the experiment ID for 'chemical_reaction', then get the results for that experiment, and tell me if the yield was above 80%. Please stream your analysis process."
+      input: "I need a complete analysis: First, get the experiment ID for 'chemical_reaction', then get the results for that experiment, and tell me if the yield was above 80%. Return only one tool call per step.  Please stream your analysis process."
      tools:
      - type: mcp
        server_label: "localmcp"
--- a/tests/integration/non_ci/responses/test_responses.py
+++ b/tests/integration/non_ci/responses/test_responses.py
@ -363,6 +363,9 @@ def test_response_non_streaming_file_search_empty_vector_store(request, compat_c
    ids=case_id_generator,
 )
 def test_response_non_streaming_mcp_tool(request, compat_client, text_model_id, case):
+    if not isinstance(compat_client, LlamaStackAsLibraryClient):
+        pytest.skip("in-process MCP server is only supported in library client")
+
    with make_mcp_server() as mcp_server_info:
        tools = case["tools"]
        for tool in tools:
@ -485,8 +488,11 @@ def test_response_non_streaming_multi_turn_image(request, compat_client, text_mo
    responses_test_cases["test_response_multi_turn_tool_execution"]["test_params"]["case"],
    ids=case_id_generator,
 )
-def test_response_non_streaming_multi_turn_tool_execution(request, compat_client, text_model_id, case):
+def test_response_non_streaming_multi_turn_tool_execution(compat_client, text_model_id, case):
    """Test multi-turn tool execution where multiple MCP tool calls are performed in sequence."""
+    if not isinstance(compat_client, LlamaStackAsLibraryClient):
+        pytest.skip("in-process MCP server is only supported in library client")
+
    with make_mcp_server(tools=dependency_tools()) as mcp_server_info:
        tools = case["tools"]
        # Replace the placeholder URL with the actual server URL
@ -541,8 +547,11 @@ def test_response_non_streaming_multi_turn_tool_execution(request, compat_client
    responses_test_cases["test_response_multi_turn_tool_execution_streaming"]["test_params"]["case"],
    ids=case_id_generator,
 )
-async def test_response_streaming_multi_turn_tool_execution(request, compat_client, text_model_id, case):
+def test_response_streaming_multi_turn_tool_execution(compat_client, text_model_id, case):
    """Test streaming multi-turn tool execution where multiple MCP tool calls are performed in sequence."""
+    if not isinstance(compat_client, LlamaStackAsLibraryClient):
+        pytest.skip("in-process MCP server is only supported in library client")
+
    with make_mcp_server(tools=dependency_tools()) as mcp_server_info:
        tools = case["tools"]
        # Replace the placeholder URL with the actual server URL
@ -634,7 +643,7 @@ async def test_response_streaming_multi_turn_tool_execution(request, compat_clie
        },
    ],
 )
-def test_response_text_format(request, compat_client, text_model_id, text_format):
+def test_response_text_format(compat_client, text_model_id, text_format):
    if isinstance(compat_client, LlamaStackAsLibraryClient):
        pytest.skip("Responses API text format is not yet supported in library client.")

@ -653,7 +662,7 @@ def test_response_text_format(request, compat_client, text_model_id, text_format


@pytest.fixture
-def vector_store_with_filtered_files(request, compat_client, text_model_id, tmp_path_factory):
+def vector_store_with_filtered_files(compat_client, text_model_id, tmp_path_factory):
    """Create a vector store with multiple files that have different attributes for filtering tests."""
    if isinstance(compat_client, LlamaStackAsLibraryClient):
        pytest.skip("Responses API file search is not yet supported in library client.")
--- a/tests/integration/vector_io/test_openai_vector_stores.py
+++ b/tests/integration/vector_io/test_openai_vector_stores.py
@ -9,10 +9,11 @@ import time
 from io import BytesIO

 import pytest
-from llama_stack_client import BadRequestError, LlamaStackClient
+from llama_stack_client import BadRequestError
 from openai import BadRequestError as OpenAIBadRequestError

 from llama_stack.apis.vector_io import Chunk
+from llama_stack.core.library_client import LlamaStackAsLibraryClient

 logger = logging.getLogger(__name__)

@ -475,9 +476,6 @@ def test_openai_vector_store_attach_file(compat_client_with_empty_stores, client
    """Test OpenAI vector store attach file."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files attach is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -526,9 +524,6 @@ def test_openai_vector_store_attach_files_on_creation(compat_client_with_empty_s
    """Test OpenAI vector store attach files on creation."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files attach is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create some files and attach them to the vector store
@ -582,9 +577,6 @@ def test_openai_vector_store_list_files(compat_client_with_empty_stores, client_
    """Test OpenAI vector store list files."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files list is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -597,16 +589,20 @@ def test_openai_vector_store_list_files(compat_client_with_empty_stores, client_
            file_buffer.name = f"openai_test_{i}.txt"
            file = compat_client.files.create(file=file_buffer, purpose="assistants")

-        compat_client.vector_stores.files.create(
+        response = compat_client.vector_stores.files.create(
            vector_store_id=vector_store.id,
            file_id=file.id,
        )
+        assert response is not None
+        assert response.status == "completed", (
+            f"Failed to attach file {file.id} to vector store {vector_store.id}: {response=}"
+        )
        file_ids.append(file.id)

    files_list = compat_client.vector_stores.files.list(vector_store_id=vector_store.id)
    assert files_list
    assert files_list.object == "list"
-    assert files_list.data
+    assert files_list.data is not None
    assert not files_list.has_more
    assert len(files_list.data) == 3
    assert set(file_ids) == {file.id for file in files_list.data}
@ -642,12 +638,13 @@ def test_openai_vector_store_list_files_invalid_vector_store(compat_client_with_
    """Test OpenAI vector store list files with invalid vector store ID."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files list is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores
+    if isinstance(compat_client, LlamaStackAsLibraryClient):
+        errors = ValueError
+    else:
+        errors = (BadRequestError, OpenAIBadRequestError)

-    with pytest.raises((BadRequestError, OpenAIBadRequestError)):
+    with pytest.raises(errors):
        compat_client.vector_stores.files.list(vector_store_id="abc123")


@ -655,9 +652,6 @@ def test_openai_vector_store_retrieve_file_contents(compat_client_with_empty_sto
    """Test OpenAI vector store retrieve file contents."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files retrieve contents is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -685,9 +679,15 @@ def test_openai_vector_store_retrieve_file_contents(compat_client_with_empty_sto
        file_id=file.id,
    )

-    assert file_contents
-    assert file_contents.content[0]["type"] == "text"
-    assert file_contents.content[0]["text"] == test_content.decode("utf-8")
+    assert file_contents is not None
+    assert len(file_contents.content) == 1
+    content = file_contents.content[0]
+
+    # llama-stack-client returns a model, openai-python is a badboy and returns a dict
+    if not isinstance(content, dict):
+        content = content.model_dump()
+    assert content["type"] == "text"
+    assert content["text"] == test_content.decode("utf-8")
    assert file_contents.filename == file_name
    assert file_contents.attributes == attributes

@ -696,9 +696,6 @@ def test_openai_vector_store_delete_file(compat_client_with_empty_stores, client
    """Test OpenAI vector store delete file."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files list is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -751,9 +748,6 @@ def test_openai_vector_store_delete_file_removes_from_vector_store(compat_client
    """Test OpenAI vector store delete file removes from vector store."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files attach is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -792,9 +786,6 @@ def test_openai_vector_store_update_file(compat_client_with_empty_stores, client
    """Test OpenAI vector store update file."""
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files update is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store
@ -840,9 +831,6 @@ def test_create_vector_store_files_duplicate_vector_store_name(compat_client_wit
    """
    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)

-    if isinstance(compat_client_with_empty_stores, LlamaStackClient):
-        pytest.skip("Vector Store Files create is not yet supported with LlamaStackClient")
-
    compat_client = compat_client_with_empty_stores

    # Create a vector store with files
--- a/uv.lock
+++ b/uv.lock
@ -476,7 +476,7 @@ wheels = [

 [[package]]
 name = "chromadb"
-version = "1.0.15"
+version = "1.0.16"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "bcrypt" },
@ -507,13 +507,13 @@ dependencies = [
    { name = "typing-extensions" },
    { name = "uvicorn", extra = ["standard"] },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ad/e2/0653b2e539db5512d2200c759f1bc7f9ef5609fe47f3c7d24b82f62dc00f/chromadb-1.0.15.tar.gz", hash = "sha256:3e910da3f5414e2204f89c7beca1650847f2bf3bd71f11a2e40aad1eb31050aa", size = 1218840, upload-time = "2025-07-02T17:07:09.875Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/15/2a/5b7e793d2a27c425e9f1813e9cb965b70e9bda08b69ee15a10e07dc3e59a/chromadb-1.0.16.tar.gz", hash = "sha256:3c864b5beb5e131bdc1f83c0b63a01ec481c6ee52028f088563ffba8478478e1", size = 1241545, upload-time = "2025-08-08T00:25:41.414Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/85/5a/866c6f0c2160cbc8dca0cf77b2fb391dcf435b32a58743da1bc1a08dc442/chromadb-1.0.15-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:51791553014297798b53df4e043e9c30f4e8bd157647971a6bb02b04bfa65f82", size = 18838820, upload-time = "2025-07-02T17:07:07.632Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/18/ff9b58ab5d334f5ecff7fdbacd6761bac467176708fa4d2500ae7c048af0/chromadb-1.0.15-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:48015803c0631c3a817befc276436dc084bb628c37fd4214047212afb2056291", size = 18057131, upload-time = "2025-07-02T17:07:05.15Z" },
-    { url = "https://files.pythonhosted.org/packages/31/49/74e34cc5aeeb25aff2c0ede6790b3671e14c1b91574dd8f98d266a4c5aad/chromadb-1.0.15-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3b73cd6fb32fcdd91c577cca16ea6112b691d72b441bb3f2140426d1e79e453a", size = 18595284, upload-time = "2025-07-02T17:06:59.102Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/33/190df917a057067e37f8b48d082d769bed8b3c0c507edefc7b6c6bb577d0/chromadb-1.0.15-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:479f1b401af9e7c20f50642ffb3376abbfd78e2b5b170429f7c79eff52e367db", size = 19526626, upload-time = "2025-07-02T17:07:02.163Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/30/6890da607358993f87a01e80bcce916b4d91515ce865f07dc06845cb472f/chromadb-1.0.15-cp39-abi3-win_amd64.whl", hash = "sha256:e0cb3b93fdc42b1786f151d413ef36299f30f783a30ce08bf0bfb12e552b4190", size = 19520490, upload-time = "2025-07-02T17:07:11.559Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/9d/bffcc814272c9b7982551803b2d45b77f39eeea1b9e965c00c05ee81c649/chromadb-1.0.16-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:144163ce7ca4f4448684d5d0c13ebb37c4d68490ecb60967a95d05cea30e0d2d", size = 18942157, upload-time = "2025-08-08T00:25:38.459Z" },
+    { url = "https://files.pythonhosted.org/packages/58/4e/de0086f3cbcfd667d75d112bb546386803ab5335599bf7099272a675e98b/chromadb-1.0.16-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:4ebcc5894e6fbb6b576452bbf4659746bfe58d9daf99a18363364e9497434bd2", size = 18147831, upload-time = "2025-08-08T00:25:35.546Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/7f/a8aff4ce96281bcb9731d10b2554f41963dd0b47acb4f90a78b2b7c4f199/chromadb-1.0.16-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:937051fc3aae94f7c171503d8f1f7662820aacc75acf45f28d3656c75c5ff1f8", size = 18682195, upload-time = "2025-08-08T00:25:29.654Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/9c/2a97d0257176aae472dff6f1ef1b7050449f384e420120e0f31d2d8f532f/chromadb-1.0.16-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0f5c5ad0c59154a9cab1506b857bab8487b588352e668cf1222c54bb9d52daa", size = 19635695, upload-time = "2025-08-08T00:25:32.68Z" },
+    { url = "https://files.pythonhosted.org/packages/96/8a/f7e810f3cbdc9186ba4e649dc32711b7ab2c23aba37cf61175f731d22293/chromadb-1.0.16-cp39-abi3-win_amd64.whl", hash = "sha256:2528c01bd8b3facca9d0e1ffac866767c386b94604df484fc792ee891c86e09a", size = 19641144, upload-time = "2025-08-08T00:25:43.446Z" },
 ]

 [[package]]
@ -1632,10 +1632,10 @@ test = [
    { name = "pypdf" },
    { name = "requests" },
    { name = "sqlalchemy", extra = ["asyncio"] },
-    { name = "torch", version = "2.7.1", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform == 'darwin'" },
-    { name = "torch", version = "2.7.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform != 'darwin'" },
-    { name = "torchvision", version = "0.22.1", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
-    { name = "torchvision", version = "0.22.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform == 'darwin'" },
+    { name = "torch", version = "2.8.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform != 'darwin'" },
+    { name = "torchvision", version = "0.23.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "torchvision", version = "0.23.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
    { name = "transformers" },
    { name = "weaviate-client" },
 ]
@ -1674,7 +1674,7 @@ requires-dist = [
    { name = "llama-api-client", specifier = ">=0.1.2" },
    { name = "llama-stack-client", specifier = ">=0.2.17" },
    { name = "llama-stack-client", marker = "extra == 'ui'", specifier = ">=0.2.17" },
-    { name = "openai", specifier = ">=1.66" },
+    { name = "openai", specifier = ">=1.99.6" },
    { name = "opentelemetry-exporter-otlp-proto-http", specifier = ">=1.30.0" },
    { name = "opentelemetry-sdk", specifier = ">=1.30.0" },
    { name = "pandas", marker = "extra == 'ui'" },
@ -2301,7 +2301,7 @@ wheels = [

 [[package]]
 name = "openai"
-version = "1.98.0"
+version = "1.99.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
    { name = "anyio" },
@ -2313,9 +2313,9 @@ dependencies = [
    { name = "tqdm" },
    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/d8/9d/52eadb15c92802711d6b6cf00df3a6d0d18b588f4c5ba5ff210c6419fc03/openai-1.98.0.tar.gz", hash = "sha256:3ee0fcc50ae95267fd22bd1ad095ba5402098f3df2162592e68109999f685427", size = 496695, upload-time = "2025-07-30T12:48:03.701Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/11/45/38a87bd6949236db5ae3132f41d5861824702b149f86d2627d6900919103/openai-1.99.6.tar.gz", hash = "sha256:f48f4239b938ef187062f3d5199a05b69711d8b600b9a9b6a3853cd271799183", size = 505364, upload-time = "2025-08-09T15:20:54.438Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a8/fe/f64631075b3d63a613c0d8ab761d5941631a470f6fa87eaaee1aa2b4ec0c/openai-1.98.0-py3-none-any.whl", hash = "sha256:b99b794ef92196829120e2df37647722104772d2a74d08305df9ced5f26eae34", size = 767713, upload-time = "2025-07-30T12:48:01.264Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/dd/9aa956485c2856346b3181542fbb0aea4e5b457fa7a523944726746da8da/openai-1.99.6-py3-none-any.whl", hash = "sha256:e40d44b2989588c45ce13819598788b77b8fb80ba2f7ae95ce90d14e46f1bd26", size = 786296, upload-time = "2025-08-09T15:20:51.95Z" },
 ]

 [[package]]
@ -4310,7 +4310,7 @@ wheels = [

 [[package]]
 name = "torch"
-version = "2.7.1"
+version = "2.8.0"
 source = { registry = "https://download.pytorch.org/whl/cpu" }
 resolution-markers = [
    "python_full_version >= '3.13' and sys_platform == 'darwin'",
@ -4326,14 +4326,14 @@ dependencies = [
    { name = "typing-extensions", marker = "sys_platform == 'darwin'" },
 ]
 wheels = [
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:7b4f8b2b83bd08f7d399025a9a7b323bdbb53d20566f1e0d584689bb92d82f9a" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:95af97e7b2cecdc89edc0558962a51921bf9c61538597dbec6b7cc48d31e2e13" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:7ecd868a086468e1bcf74b91db425c1c2951a9cfcd0592c4c73377b7e42485ae" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:a47b7986bee3f61ad217d8a8ce24605809ab425baf349f97de758815edd2ef54" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:fbe2e149c5174ef90d29a5f84a554dfaf28e003cb4f61fa2c8c024c17ec7ca58" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:057efd30a6778d2ee5e2374cd63a63f63311aa6f33321e627c655df60abdd390" },
 ]

 [[package]]
 name = "torch"
-version = "2.7.1+cpu"
+version = "2.8.0+cpu"
 source = { registry = "https://download.pytorch.org/whl/cpu" }
 resolution-markers = [
    "(python_full_version >= '3.13' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform != 'darwin' and sys_platform != 'linux')",
@ -4351,21 +4351,24 @@ dependencies = [
    { name = "typing-extensions", marker = "sys_platform != 'darwin'" },
 ]
 wheels = [
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:3bf2db5adf77b433844f080887ade049c4705ddf9fe1a32023ff84ff735aa5ad" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:8f8b3cfc53010a4b4a3c7ecb88c212e9decc4f5eeb6af75c3c803937d2d60947" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp312-cp312-win_amd64.whl", hash = "sha256:0bc887068772233f532b51a3e8c8cfc682ae62bef74bf4e0c53526c8b9e4138f" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp312-cp312-win_arm64.whl", hash = "sha256:a2618775f32eb4126c5b2050686da52001a08cffa331637d9cf51c8250931e00" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:eb17646792ac4374ffc87e42369f45d21eff17c790868963b90483ef0b6db4ef" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:84ea1f6a1d15663037d01b121d6e33bb9da3c90af8e069e5072c30f413455a57" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313-win_amd64.whl", hash = "sha256:b66f77f6f67317344ee083aa7ac4751a14395fcb38060d564bf513978d267153" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:56136a2aca6707df3c8811e46ea2d379eaafd18e656e2fd51e8e4d0ca995651b" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:355614185a2aea7155f9c88a20bfd49de5f3063866f3cf9b2f21b6e9e59e31e0" },
-    { url = "https://download.pytorch.org/whl/cpu/torch-2.7.1%2Bcpu-cp313-cp313t-win_amd64.whl", hash = "sha256:464bca1bc9452f2ccd676514688896e66b9488f2a0268ecd3ac497cf09c5aac1" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp312-cp312-linux_s390x.whl", hash = "sha256:0e34e276722ab7dd0dffa9e12fe2135a9b34a0e300c456ed7ad6430229404eb5" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:610f600c102386e581327d5efc18c0d6edecb9820b4140d26163354a99cd800d" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:cb9a8ba8137ab24e36bf1742cb79a1294bd374db570f09fc15a5e1318160db4e" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp312-cp312-win_amd64.whl", hash = "sha256:2be20b2c05a0cce10430cc25f32b689259640d273232b2de357c35729132256d" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp312-cp312-win_arm64.whl", hash = "sha256:99fc421a5d234580e45957a7b02effbf3e1c884a5dd077afc85352c77bf41434" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313-linux_s390x.whl", hash = "sha256:8b5882276633cf91fe3d2d7246c743b94d44a7e660b27f1308007fdb1bb89f7d" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:a5064b5e23772c8d164068cc7c12e01a75faf7b948ecd95a0d4007d7487e5f25" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:8f81dedb4c6076ec325acc3b47525f9c550e5284a18eae1d9061c543f7b6e7de" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313-win_amd64.whl", hash = "sha256:e1ee1b2346ade3ea90306dfbec7e8ff17bc220d344109d189ae09078333b0856" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313-win_arm64.whl", hash = "sha256:64c187345509f2b1bb334feed4666e2c781ca381874bde589182f81247e61f88" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:af81283ac671f434b1b25c95ba295f270e72db1fad48831eb5e4748ff9840041" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:a9dbb6f64f63258bc811e2c0c99640a81e5af93c531ad96e95c5ec777ea46dab" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.8.0%2Bcpu-cp313-cp313t-win_amd64.whl", hash = "sha256:6d93a7165419bc4b2b907e859ccab0dea5deeab261448ae9a5ec5431f14c0e64" },
 ]

 [[package]]
 name = "torchvision"
-version = "0.22.1"
+version = "0.23.0"
 source = { registry = "https://download.pytorch.org/whl/cpu" }
 resolution-markers = [
    "python_full_version >= '3.13' and platform_machine == 'aarch64' and sys_platform == 'linux'",
@ -4376,21 +4379,21 @@ resolution-markers = [
 dependencies = [
    { name = "numpy", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
    { name = "pillow", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
-    { name = "torch", version = "2.7.1", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform == 'darwin'" },
-    { name = "torch", version = "2.7.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine == 'aarch64' and sys_platform == 'linux'" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "sys_platform == 'darwin'" },
+    { name = "torch", version = "2.8.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "platform_machine == 'aarch64' and sys_platform == 'linux'" },
 ]
 wheels = [
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:153f1790e505bd6da123e21eee6e83e2e155df05c0fe7d56347303067d8543c5" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:964414eef19459d55a10e886e2fca50677550e243586d1678f65e3f6f6bac47a" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9c3ae3319624c43cc8127020f46c14aa878406781f0899bb6283ae474afeafbf" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:4a614a6a408d2ed74208d0ea6c28a2fbb68290e9a7df206c5fef3f0b6865d307" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:043d9e35ed69c2e586aff6eb9e2887382e7863707115668ac9d140da58f42cba" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:27142bcc8a984227a6dcf560985e83f52b82a7d3f5fe9051af586a2ccc46ef26" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:e0e2c04a91403e8dd3af9756c6a024a1d9c0ed9c0d592a8314ded8f4fe30d440" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:6dd7c4d329a0e03157803031bc856220c6155ef08c26d4f5bbac938acecf0948" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1c37e325e09a184b730c3ef51424f383ec5745378dc0eca244520aca29722600" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:2f7fd6c15f3697e80627b77934f77705f3bc0e98278b989b2655de01f6903e1d" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:2df618e1143805a7673aaf82cb5720dd9112d4e771983156aaf2ffff692eebf9" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:2a3299d2b1d5a7aed2d3b6ffb69c672ca8830671967eb1cee1497bacd82fe47b" },
 ]

 [[package]]
 name = "torchvision"
-version = "0.22.1+cpu"
+version = "0.23.0+cpu"
 source = { registry = "https://download.pytorch.org/whl/cpu" }
 resolution-markers = [
    "(python_full_version >= '3.13' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform != 'darwin' and sys_platform != 'linux')",
@ -4399,15 +4402,15 @@ resolution-markers = [
 dependencies = [
    { name = "numpy", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
    { name = "pillow", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
-    { name = "torch", version = "2.7.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "torch", version = "2.8.0+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
 ]
 wheels = [
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:b5fa7044bd82c6358e8229351c98070cf3a7bf4a6e89ea46352ae6c65745ef94" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp312-cp312-win_amd64.whl", hash = "sha256:433cb4dbced7291f17064cea08ac1e5aebd02ec190e1c207d117ad62a8961f2b" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:a93c21f18c33a819616b3dda7655aa4de40b219682c654175b6bbeb65ecc2e5f" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp313-cp313-win_amd64.whl", hash = "sha256:34c914ad4728b81848ac802c5fc5eeb8de8ff4058cc59c1463a74ce4f4fbf0d8" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:ab7ae82529887c704c1b5d1d5198f65dc777d04fc3858b374503a6deedb82b19" },
-    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.22.1%2Bcpu-cp313-cp313t-win_amd64.whl", hash = "sha256:b2d1c4bdbfd8e6c779dc810a6171b56224f1332fc46986810d4081bed1633804" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:ae459d4509d3b837b978dc6c66106601f916b6d2cda75c137e3f5f48324ce1da" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp312-cp312-win_amd64.whl", hash = "sha256:a651ccc540cf4c87eb988730c59c2220c52b57adc276f044e7efb9830fa65a1d" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:dea90a67d60a5366b0358a0b8d6bf267805278697d6fd950cf0e31139e56d1be" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp313-cp313-win_amd64.whl", hash = "sha256:82928788025170c62e7df1120dcdc0cd175bfc31c08374613ce6d1a040bc0cda" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:474d77adbbbed5166db3e5636b4b4ae3399c66ef5bfa12536e254b32259c90c0" },
+    { url = "https://download.pytorch.org/whl/cpu/torchvision-0.23.0%2Bcpu-cp313-cp313t-win_amd64.whl", hash = "sha256:8d6a47e23d7896f0ef9aa7ea7179eb6324e82438aa66d19884c2020d0646b104" },
 ]

 [[package]]