# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
- This PR implements keyword and hybrid search for Weaviate DB based on
its inbuilt functions.
- Added fixtures to conftest.py for Weaviate.
- Enabled integration tests for remote Weaviate on all 3 search modes.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
Closes#3010
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
Unit tests and integration tests should pass on this PR.
# What does this PR do?
add ModelsProtocolPrivate methods to OpenAIMixin
this will allow providers using OpenAIMixin to use a common interface
## Test Plan
ci w/ new tests
# What does this PR do?
closes#3268closes#3498
When resuming from previous response ID, currently we attempt to convert
from the stored responses input to chat completion messages, which is
not always possible, e.g. for tool calls where some data is lost once
converted from chat completion message to repsonses input format.
This PR stores the chat completion messages that correspond to the
_last_ call to chat completion, which is sufficient to be resumed from
in the next responses API call, where we load these saved messages and
skip conversion entirely.
Separate issue to optimize storage:
https://github.com/llamastack/llama-stack/issues/3646
## Test Plan
existing CI tests
This is a sweeping change to clean up some gunk around our "Tool"
definitions.
First, we had two types `Tool` and `ToolDef`. The first of these was a
"Resource" type for the registry but we had stopped registering tools
inside the Registry long back (and only registered ToolGroups.) The
latter was for specifying tools for the Agents API. This PR removes the
former and adds an optional `toolgroup_id` field to the latter.
Secondly, as pointed out by @bbrowning in
https://github.com/llamastack/llama-stack/pull/3003#issuecomment-3245270132,
we were doing a lossy conversion from a full JSON schema from the MCP
tool specification into our ToolDefinition to send it to the model.
There is no necessity to do this -- we ourselves aren't doing any
execution at all but merely passing it to the chat completions API which
supports this. By doing this (and by doing it poorly), we encountered
limitations like not supporting array items, or not resolving $refs,
etc.
To fix this, we replaced the `parameters` field by `{ input_schema,
output_schema }` which can be full blown JSON schemas.
Finally, there were some types in our llama-related chat format
conversion which needed some cleanup. We are taking this opportunity to
clean those up.
This PR is a substantial breaking change to the API. However, given our
window for introducing breaking changes, this suits us just fine. I will
be landing a concurrent `llama-stack-client` change as well since API
shapes are changing.
# What does this PR do?
the LiteLLMOpenAIMixin provides support for reading key from provider
data (headers users send).
this adds the same functionality to the OpenAIMixin.
this is infrastructure for migrating providers.
## Test Plan
ci w/ new tests
# What does this PR do?
When a model decides to use an MCP tool call that requires no arguments,
it sets the `arguments` field to `None`. This causes the user to see a
`400 bad requst error` due to validation errors down the stack because
this field gets removed when being parsed by an openai compatible
inference provider like vLLM
This PR ensures that, as soon as the tool call args are accumulated
while streaming, we check to ensure no tool call function arguments are
set to None - if they are we replace them with "{}"
<!-- If resolving an issue, uncomment and update the line below -->
Closes#3456
## Test Plan
Added new unit test to verify that any tool calls with function
arguments set to `None` get handled correctly
---------
Signed-off-by: Jaideep Rao <jrao@redhat.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
# What does this PR do?
Fireworks doesn't allow repsonse_format with tool use. The default
response format is 'text' anyway, so we can safely omit.
## Test Plan
Below script failed without the change, runs after.
```
#!/usr/bin/env python3
"""
Script to test Responses API with kubernetes-mcp-server.
This script:
1. Connects to the llama stack server
2. Uses the Responses API with MCP tools
3. Asks for the list of Kubernetes namespaces using the kubernetes-mcp-server
"""
import json
from openai import OpenAI
# Connect to the llama stack server
base_url = "http://localhost:8321/v1"
client = OpenAI(base_url=base_url, api_key="fake")
# Define the MCP tool pointing to the kubernetes-mcp-server
# The kubernetes-mcp-server is running on port 3000 with SSE endpoint at /sse
mcp_server_url = "http://localhost:3000/sse"
tools = [
{
"type": "mcp",
"server_label": "k8s",
"server_url": mcp_server_url,
}
]
# Create a response request asking for k8s namespaces
print("Sending request to list Kubernetes namespaces...")
print(f"Using MCP server at: {mcp_server_url}")
print("Available tools will be listed automatically by the MCP server.")
print()
response = client.responses.create(
# model="meta-llama/Llama-3.2-3B-Instruct", # Using the vllm model
model="fireworks/accounts/fireworks/models/llama4-scout-instruct-basic",
# model="openai/gpt-4o",
input="what are all the Kubernetes namespaces? Use tool call to `namespaces_list`. make sure to adhere to the tool calling format UNDER ALL CIRCUMSTANCES.",
tools=tools,
stream=False,
)
print("\n" + "=" * 80)
print("RESPONSE OUTPUT:")
print("=" * 80)
# Print the output
for i, output in enumerate(response.output):
print(f"\n[Output {i + 1}] Type: {output.type}")
if output.type == "mcp_list_tools":
print(f" Server: {output.server_label}")
print(f" Tools available: {[t.name for t in output.tools]}")
elif output.type == "mcp_call":
print(f" Tool called: {output.name}")
print(f" Arguments: {output.arguments}")
print(f" Result: {output.output}")
if output.error:
print(f" Error: {output.error}")
elif output.type == "message":
print(f" Role: {output.role}")
print(f" Content: {output.content}")
print("\n" + "=" * 80)
print("FINAL RESPONSE TEXT:")
print("=" * 80)
print(response.output_text)
```
https://github.com/llamastack/llama-stack/pull/3604 broke multipart form
data field parsing for the Files API since it changed its shape -- so as
to match the API exactly to the OpenAI spec even in the generated client
code.
The underlying reason is that multipart/form-data cannot transport
structured nested fields. Each field must be str-serialized. The client
(specifically the OpenAI client whose behavior we must match),
transports sub-fields as `expires_after[anchor]` and
`expires_after[seconds]`, etc. We must be able to handle these fields
somehow on the server without compromising the shape of the YAML spec.
This PR "fixes" this by adding a dependency to convert the data. The
main trade-off here is that we must add this `Depends()` annotation on
every provider implementation for Files. This is a headache, but a much
more reasonable one (in my opinion) given the alternatives.
## Test Plan
Tests as shown in
https://github.com/llamastack/llama-stack/pull/3604#issuecomment-3351090653
pass.
# What does this PR do?
Fixes error:
```
[ERROR] Error executing endpoint route='/v1/openai/v1/responses'
method='post': Error code: 400 - {'error': {'message': "Invalid schema for function 'pods_exec': In context=('properties', 'command'), array
schema missing items.", 'type': 'invalid_request_error', 'param': 'tools[7].function.parameters', 'code': 'invalid_function_parameters'}}
```
From script:
```
#!/usr/bin/env python3
"""
Script to test Responses API with kubernetes-mcp-server.
This script:
1. Connects to the llama stack server
2. Uses the Responses API with MCP tools
3. Asks for the list of Kubernetes namespaces using the kubernetes-mcp-server
"""
import json
from openai import OpenAI
# Connect to the llama stack server
base_url = "http://localhost:8321/v1/openai/v1"
client = OpenAI(base_url=base_url, api_key="fake")
# Define the MCP tool pointing to the kubernetes-mcp-server
# The kubernetes-mcp-server is running on port 3000 with SSE endpoint at /sse
mcp_server_url = "http://localhost:3000/sse"
tools = [
{
"type": "mcp",
"server_label": "k8s",
"server_url": mcp_server_url,
}
]
# Create a response request asking for k8s namespaces
print("Sending request to list Kubernetes namespaces...")
print(f"Using MCP server at: {mcp_server_url}")
print("Available tools will be listed automatically by the MCP server.")
print()
response = client.responses.create(
# model="meta-llama/Llama-3.2-3B-Instruct", # Using the vllm model
model="openai/gpt-4o",
input="what are all the Kubernetes namespaces? Use tool call to `namespaces_list`. make sure to adhere to the tool calling format.",
tools=tools,
stream=False,
)
print("\n" + "=" * 80)
print("RESPONSE OUTPUT:")
print("=" * 80)
# Print the output
for i, output in enumerate(response.output):
print(f"\n[Output {i + 1}] Type: {output.type}")
if output.type == "mcp_list_tools":
print(f" Server: {output.server_label}")
print(f" Tools available: {[t.name for t in output.tools]}")
elif output.type == "mcp_call":
print(f" Tool called: {output.name}")
print(f" Arguments: {output.arguments}")
print(f" Result: {output.output}")
if output.error:
print(f" Error: {output.error}")
elif output.type == "message":
print(f" Role: {output.role}")
print(f" Content: {output.content}")
print("\n" + "=" * 80)
print("FINAL RESPONSE TEXT:")
print("=" * 80)
print(response.output_text)
```
## Test Plan
new unit tests
script now runs successfully
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
This PR fix#3300 by adding mime type of application/json support in
[agent_instance.py](4a59961a6c/llama_stack/providers/inline/agents/meta_reference/agent_instance.py (L923))
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[3300] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
all related pytest passed, see log:
```
./scripts/unit-tests.sh tests/unit/providers/agent/test_get_raw_document_text.py -vvv
/Users/kaiwu/work/kaiwu/llama-stack/.venv/bin/python3
Uninstalled 22 packages in 5.65s
Installed 47 packages in 1.24s
================= test session starts =================
platform darwin -- Python 3.12.9, pytest-8.4.2, pluggy-1.6.0 -- /Users/kaiwu/work/kaiwu/llama-stack/.venv/bin/python
cachedir: .pytest_cache
metadata: {'Python': '3.12.9', 'Platform': 'macOS-15.6.1-arm64-arm-64bit', 'Packages': {'pytest': '8.4.2', 'pluggy': '1.6.0'}, 'Plugins': {'anyio': '4.9.0', 'html': '4.1.1', 'socket': '0.7.0', 'asyncio': '1.1.0', 'json-report': '1.5.0', 'timeout': '2.4.0', 'metadata': '3.1.1', 'cov': '6.2.1', 'nbval': '0.11.0'}}
rootdir: /Users/kaiwu/work/kaiwu/llama-stack
configfile: pyproject.toml
plugins: anyio-4.9.0, html-4.1.1, socket-0.7.0, asyncio-1.1.0, json-report-1.5.0, timeout-2.4.0, metadata-3.1.1, cov-6.2.1, nbval-0.11.0
asyncio: mode=Mode.AUTO, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
collected 14 items
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_text_mime_types PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_yaml_mime_type PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_deprecated_text_yaml_with_warning PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_deprecated_text_yaml_with_url PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_deprecated_text_yaml_with_text_content_item PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_json_mime_type PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_json_url PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_json_text_content_item PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_rejects_unsupported_mime_types PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_url_content PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_yaml_url PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_text_content_item PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_yaml_text_content_item PASSED
tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_rejects_unexpected_content_type PASSED
================ slowest 10 durations =================
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_deprecated_text_yaml_with_url
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_rejects_unsupported_mime_types
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_rejects_unexpected_content_type
0.00s setup tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_text_mime_types
0.00s teardown tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_text_mime_types
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_yaml_url
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_url_content
0.00s teardown tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_rejects_unsupported_mime_types
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_with_json_url
0.00s call tests/unit/providers/agent/test_get_raw_document_text.py::test_get_raw_document_text_supports_text_mime_types
================= 14 passed in 0.14s ==================
Generating coverage report...
Wrote HTML report to htmlcov-3.12/index.html
```
# What does this PR do?
Mirroring the same changes that was used for inference_store:
https://github.com/llamastack/llama-stack/pull/3383
Will follow up with a shared internal API for managing these write
queues.
## Test Plan
existing tests
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
Add items and title to ToolParameter/ToolParamDefinition. Adding items
will resolve the issue that occurs with Gemini LLM when an MCP tool has
array-type properties.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
Unite test cases will be added.
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Kai Wu <kaiwu@meta.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
# What does this PR do?
simplify Ollama inference adapter by -
- moving image_url download code to OpenAIMixin
- being a ModelRegistryHelper instead of having one (mypy blocks
check_model_availability method assignment)
## Test Plan
- add unit tests for new download feature
- add integration tests for openai_chat_completion w/ image_url (close
test gap)
the commit to be reverted is an public api behavior change to something
we should not support.
instead of allowing silent updates (the caller cannot see the log
messages), we should be sending an error to the caller that they must
first unregister the model before reusing the same name w/ a different
backend.
# What does this PR do?
When listing (and lazily indexing) tools, it's possible for an error to
get thrown by individual toolgroups if for example an MCP toolgroup is
unable to connect to its `mcp_endpoint`.
This logs a warning in the server when that happens, logs a full stack
trace of the error if debug logging is enabled, and just returns the
list of tools from all working toolgroups instead of throwing an error
to the client when a single toolgroup is temporarily or permanently
misbehaving.
The exception to the above is authentication errors, which we
specifically send all the way back to the client as that's how we
indicate to the client that it needs to provide authentication data for
the remote MCP servers.
Closes#2540
## Test Plan
A new unit test was added to test this exception handling, which is run
as part of our regular test suite but also manually run to specifically
verify this fix via:
```
uv run pytest -sv --asyncio-mode=auto \
tests/unit/distribution/routers/test_routing_tables.py
```
To verify the additional debug logging is printing properly:
```
LLAMA_STACK_LOGGING=core=debug \
uv run pytest -sv --asyncio-mode=auto \
tests/unit/distribution/routers/test_routing_tables.py
```
The mcp integration tests were run as below (and by CI):
```
ollama run llama3.2:3b
ENABLE_OLLAMA="ollama" \
OLLAMA_INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct" \
LLAMA_STACK_CONFIG=starter \
uv run pytest -sv tests/integration/tool_runtime/test_mcp.py \
--text-model meta-llama/Llama-3.2-3B-Instruct
```
---------
Signed-off-by: Ben Browning <bbrownin@redhat.com>
Signed-off-by: Sébastien Han <seb@redhat.com>
Co-authored-by: Sébastien Han <seb@redhat.com>
# What does this PR do?
- remove auto-download of ollama embedding models
- add embedding model metadata to dynamic listing w/ unit test
- add support and tests for allowed_models
- removed inference provider models.py files where dynamic listing is
enabled
- store embedding metadata in embedding_model_metadata field on
inference providers
- make model_entries optional on ModelRegistryHelper and
LiteLLMOpenAIMixin
- make OpenAIMixin a ModelRegistryHelper
- skip base64 embedding test for remote::ollama, always returns floats
- only use OpenAI client for ollama model listing
- remove unused build_model_entry function
- remove unused get_huggingface_repo function
## Test Plan
ci w/ new tests
# What does this PR do?
This PR is generated with AI and reviewed by me.
Refactors the AuthorizedSqlStore class to store the access policy as an
instance variable rather than passing it as a parameter to each method
call. This simplifies the API.
# Test Plan
existing tests
# What does this PR do?
As shown in #3421, we can scale stack to handle more RPS with k8s
replicas. This PR enables multi process stack with uvicorn --workers so
that we can achieve the same scaling without being in k8s.
To achieve that we refactor main to split out the app construction
logic. This method needs to be non-async. We created a new `Stack` class
to house impls and have a `start()` method to be called in lifespan to
start background tasks instead of starting them in the old
`construct_stack`. This way we avoid having to manage an event loop
manually.
## Test Plan
CI
> uv run --with llama-stack python -m llama_stack.core.server.server
benchmarking/k8s-benchmark/stack_run_config.yaml
works.
> LLAMA_STACK_CONFIG=benchmarking/k8s-benchmark/stack_run_config.yaml uv
run uvicorn llama_stack.core.server.server:create_app --port 8321
--workers 4
works.
# What does this PR do?
currently `RemoteProviderSpec` has an `AdapterSpec` embedded in it.
Remove `AdapterSpec`, and put its leftover fields into
`RemoteProviderSpec`.
Additionally, many of the fields were duplicated between
`InlineProviderSpec` and `RemoteProviderSpec`. Move these to
`ProviderSpec` so they are shared.
Fixup the distro codegen to use `RemoteProviderSpec` directly rather
than `remote_provider_spec` which took an AdapterSpec and returned a
full provider spec
## Test Plan
existing distro tests should pass.
Signed-off-by: Charlie Doern <cdoern@redhat.com>
# What does this PR do?
this replaces the static model listing for any provider using
OpenAIMixin
currently -
- anthropic
- azure openai
- gemini
- groq
- llama-api
- nvidia
- openai
- sambanova
- tgi
- vertexai
- vllm
- not changed: together has its own impl
## Test Plan
- new unit tests
- manual for llama-api, openai, groq, gemini
```
for provider in llama-openai-compat openai groq gemini; do
uv run llama stack build --image-type venv --providers inference=remote::provider --run &
uv run --with llama-stack-client llama-stack-client models list | grep Total
```
results (17 sep 2025):
- llama-api: 4
- openai: 86
- groq: 21
- gemini: 66
closes#3467
# What does this PR do?
*Add dynamic authentication token forwarding support for vLLM provider*
This enables per-request authentication tokens for vLLM providers,
supporting use cases like RAG operations where different requests may
need different authentication tokens. The implementation follows the
same pattern as other providers like Together AI, Fireworks, and
Passthrough.
- Add LiteLLMOpenAIMixin that manages the vllm_api_token properly
Usage:
- Static: VLLM_API_TOKEN env var or config.api_token
- Dynamic: X-LlamaStack-Provider-Data header with vllm_api_token
All existing functionality is preserved while adding new dynamic
capabilities.
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
```
curl -X POST "http://localhost:8000/v1/chat/completions" -H "Authorization: Bearer my-dynamic-token" \
-H "X-LlamaStack-Provider-Data: {\"vllm_api_token\": \"Bearer my-dynamic-token\", \"vllm_url\": \"http://dynamic-server:8000\"}" \
-H "Content-Type: application/json" \
-d '{"model": "llama-3.1-8b", "messages": [{"role": "user", "content": "Hello!"}]}'
```
---------
Signed-off-by: Akram Ben Aissi <akram.benaissi@gmail.com>
# What does this PR do?
Fixes this warning in llama stack build:
```bash
WARNING 2025-09-15 15:29:02,197 llama_stack.core.distribution:149 core: Failed to import module prompts: No module named
'llama_stack.providers.registry.prompts'"
```
## Test Plan
Test added
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
# What does this PR do?
Modified the code in registry.py.
The key changes are:
1. Removed the `return False` statement
2. Added a warning log message that includes the object type,
identifier, and provider_id for better debugging.
3. The method now continues with the registration process instead of
early returning.
---------
Co-authored-by: Omar Abdelwahab <omara@fb.com>
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
This PR provides functionality for users to unregister ScoringFn and
Benchmark resources for `scoring` and `eval` APIs.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
Closes#3051
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
Updated integration and unit tests via CI workflow
# What does this PR do?
the @required_args decorator in openai-python is masking the async
nature of the {AsyncCompletions,chat.AsyncCompletions}.create method.
see https://github.com/openai/openai-python/issues/996
this means two things -
0. we cannot use iscoroutine in the recorder to detect async vs non
1. our mocks are inappropriately introducing identifiable async
for (0), we update the iscoroutine check w/ detection of /v1/models,
which is the only non-async function we mock & record.
for (1), we could leave everything as is and assume (0) will catch
errors. to be defensive, we update the unit tests to mock below create
methods, allowing the true openai-python create() methods to be tested.
# What does this PR do?
the recorder mocks the openai-python interface. the openai-python
interface allows NOT_GIVEN as an input option. this change properly
handles NOT_GIVEN.
## Test Plan
ci (coverage for chat, completions, embeddings)
# What does this PR do?
Migrates MD5 and SHA-1 hash algorithms to SHA-256.
In particular, replaces:
- MD5 in chunk ID generation.
- MD5 in file verification.
- SHA-1 in model identifier digests.
And updates all related test expectations.
Original discussion:
https://github.com/llamastack/llama-stack/discussions/3413
<!-- If resolving an issue, uncomment and update the line below -->
Closes#3424.
## Test Plan
Unit tests from scripts/unit-tests.sh were updated to match the new hash
output, and ran to verify the tests pass.
Signed-off-by: Doug Edgar <dedgar@redhat.com>
# What does this PR do?
update the async detection test for vllm
- remove a network access from unit tests
- remove direct logging use
the idea behind the test is to mock inference w/ a sleep, initiate
concurrent inference calls, verify the total execution time is close to
the sleep time. in a non-async env the total time would be closer to
sleep * num concurrent calls.
## Test Plan
ci
# What does this PR do?
update vLLM inference provider to use OpenAIMixin for openai-compat
functions
inference recordings from Qwen3-0.6B and vLLM 0.8.3 -
```
docker run --gpus all -v ~/.cache/huggingface:/root/.cache/huggingface -p 8000:8000 --ipc=host \
vllm/vllm-openai:latest \
--model Qwen/Qwen3-0.6B --enable-auto-tool-choice --tool-call-parser hermes
```
## Test Plan
```
./scripts/integration-tests.sh --stack-config server:ci-tests --setup vllm --subdirs inference
```
# What does this PR do?
- Updating documentation on migration from RAG Tool to Vector Stores and
Files APIs
- Adding exception handling for Vector Stores in RAG Tool
- Add more tests on migration from RAG Tool to Vector Stores
- Migrate off of inference_api for context_retriever for RAG
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
Integration and unit tests added
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Fixes#3370
AWS switched to requiring region-prefixed inference profile IDs instead
of foundation model IDs for on-demand throughput. This was causing
ValidationException errors.
Added auto-detection based on boto3 client region to convert model IDs
like meta.llama3-1-70b-instruct-v1:0 to
us.meta.llama3-1-70b-instruct-v1:0 depending on the detected region.
Also handles edge cases like ARNs, case insensitive regions, and None
regions.
Tested with this request.
```json
{
"model_id": "meta.llama3-1-8b-instruct-v1:0",
"messages": [
{
"role": "system",
"content": "You are a helpful assistant."
},
{
"role": "user",
"content": "tell me a riddle"
}
],
"sampling_params": {
"strategy": {
"type": "top_p",
"temperature": 0.7,
"top_p": 0.9
},
"max_tokens": 512
}
}
```
<img width="1488" height="878" alt="image"
src="https://github.com/user-attachments/assets/0d61beec-3869-4a31-8f37-9f554c280b88"
/>
# What does this PR do?
Adds a write worker queue for writes to inference store. This avoids
overwhelming request processing with slow inference writes.
## Test Plan
Benchmark:
```
cd /docs/source/distributions/k8s-benchmark
# start mock server
python openai-mock-server.py --port 8000
# start stack server
LLAMA_STACK_LOGGING="all=WARNING" uv run --with llama-stack python -m llama_stack.core.server.server docs/source/distributions/k8s-benchmark/stack_run_config.yaml
# run benchmark script
uv run python3 benchmark.py --duration 120 --concurrent 50 --base-url=http://localhost:8321/v1/openai/v1 --model=vllm-inference/meta-llama/Llama-3.2-3B-Instruct
```
## RPS from 21 -> 57
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
Fix pre-commit issues: non executable shebang file, @pytest.mark.asyncio
decorator
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
The test_query_adds_vector_db_id_to_chunk_metadata test was failing
because MemoryToolRuntimeImpl.__init__() now requires a files_api
parameter.
Fixes failing unit tests for Python 3.12 and 3.13.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
# What does this PR do?
When running RAG in a multi vector DB setting, it can be difficult to
trace where retrieved chunks originate from. This PR adds the
`vector_db_id` into each chunk’s metadata, making it easier to
understand which database a given chunk came from. This is helpful for
debugging and for analyzing retrieval behavior of multiple DBs.
Relevant code:
```python
for vector_db_id, result in zip(vector_db_ids, results):
for chunk, score in zip(result.chunks, result.scores):
if not hasattr(chunk, "metadata") or chunk.metadata is None:
chunk.metadata = {}
chunk.metadata["vector_db_id"] = vector_db_id
chunks.append(chunk)
scores.append(score)
```
## Test Plan
* Ran Llama Stack in debug mode.
* Verified that `vector_db_id` was added to each chunk’s metadata.
* Confirmed that the metadata was printed in the console when using the
RAG tool.
---------
Co-authored-by: are-ces <cpompeia@redhat.com>
Co-authored-by: Francisco Arceo <arceofrancisco@gmail.com>
# What does this PR do?
This PR adds support for OpenAI Prompts API.
Note, OpenAI does not explicitly expose the Prompts API but instead
makes it available in the Responses API and in the [Prompts
Dashboard](https://platform.openai.com/docs/guides/prompting#create-a-prompt).
I have added the following APIs:
- CREATE
- GET
- LIST
- UPDATE
- Set Default Version
The Set Default Version API is made available only in the Prompts
Dashboard and configures which prompt version is returned in the GET
(the latest version is the default).
Overall, the expected functionality in Responses will look like this:
```python
from openai import OpenAI
client = OpenAI()
response = client.responses.create(
prompt={
"id": "pmpt_68b0c29740048196bd3a6e6ac3c4d0e20ed9a13f0d15bf5e",
"version": "2",
"variables": {
"city": "San Francisco",
"age": 30,
}
}
)
```
### Resolves https://github.com/llamastack/llama-stack/issues/3276
## Test Plan
Unit tests added. Integration tests can be added after client
generation.
## Next Steps
1. Update Responses API to support Prompt API
2. I'll enhance the UI to implement the Prompt Dashboard.
3. Add cache for lower latency
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
# What does this PR do?
Add Kubernetes authentication provider support
- Add KubernetesAuthProvider class for token validation using Kubernetes
SelfSubjectReview API
- Add KubernetesAuthProviderConfig with configurable API server URL, TLS
settings, and claims mapping
- Implement authentication via POST requests to
/apis/authentication.k8s.io/v1/selfsubjectreviews endpoint
- Add support for parsing Kubernetes SelfSubjectReview response format
to extract user information
- Add KUBERNETES provider type to AuthProviderType enum
- Update create_auth_provider factory function to handle 'kubernetes'
provider type
- Add comprehensive unit tests for KubernetesAuthProvider functionality
- Add documentation with configuration examples and usage instructions
The provider validates tokens by sending SelfSubjectReview requests to
the Kubernetes API server and extracts user information from the
userInfo structure in the response.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
<!-- Describe the tests you ran to verify your changes with result
summaries. *Provide clear instructions so the plan can be easily
re-executed.* -->
What This Verifies:
Authentication header validation
Token validation with Kubernetes SelfSubjectReview and kubernetes server
API endpoint
Error handling for invalid tokens and HTTP errors
Request payload structure and headers
```
python -m pytest tests/unit/server/test_auth.py -k "kubernetes" -v
```
Signed-off-by: Akram Ben Aissi <akram.benaissi@gmail.com>
# What does this PR do?
update Groq inference provider to use OpenAIMixin for openai-compat
endpoints
changes on api.groq.com -
- json_schema is now supported for specific models, see
https://console.groq.com/docs/structured-outputs#supported-models
- response_format with streaming is now supported for models that
support response_format
- groq no longer returns a 400 error if tools are provided and
tool_choice is not "required"
## Test Plan
```
$ GROQ_API_KEY=... uv run llama stack build --image-type venv --providers inference=remote::groq --run
...
$ LLAMA_STACK_CONFIG=http://localhost:8321 uv run --group test pytest -v -ra --text-model groq/llama-3.3-70b-versatile tests/integration/inference/test_openai_completion.py -k 'not store'
...
SKIPPED [3] tests/integration/inference/test_openai_completion.py:44: Model groq/llama-3.3-70b-versatile hosted by remote::groq doesn't support OpenAI completions.
SKIPPED [3] tests/integration/inference/test_openai_completion.py:94: Model groq/llama-3.3-70b-versatile hosted by remote::groq doesn't support vllm extra_body parameters.
SKIPPED [4] tests/integration/inference/test_openai_completion.py:73: Model groq/llama-3.3-70b-versatile hosted by remote::groq doesn't support n param.
SKIPPED [1] tests/integration/inference/test_openai_completion.py💯 Model groq/llama-3.3-70b-versatile hosted by remote::groq doesn't support chat completion calls with base64 encoded files.
======================= 8 passed, 11 skipped, 8 deselected, 2 warnings in 5.13s ========================
```
---------
Co-authored-by: raghotham <rsm@meta.com>
# What does this PR do?
This change migrates the VectorDB id generation to Vector Stores.
This is a breaking change for **_some users_** that may have application
code using the `vector_db_id` parameter in the request of the VectorDB
protocol instead of the `VectorDB.identifier` in the response.
By default we will now create a Vector Store every time we register a
VectorDB. The caveat with this approach is that this maps the
`vector_db_id` → `vector_store.name`. This is a reasonable tradeoff to
transition users towards OpenAI Vector Stores.
As an added benefit, registering VectorDBs will result in them appearing
in the VectorStores admin UI.
### Why?
This PR makes the `POST` API call to `/v1/vector-dbs` swap the
`vector_db_id` parameter in the **request body** into the VectorStore's
name field and sets the `vector_db_id` to the generated vector store id
(e.g., `vs_038247dd-4bbb-4dbb-a6be-d5ecfd46cfdb`).
That means that users would have to do something like follows in their
application code:
```python
res = client.vector_dbs.register(
vector_db_id='my-vector-db-id',
embedding_model='ollama/all-minilm:l6-v2',
embedding_dimension=384,
)
vector_db_id = res.identifier
```
And then the rest of their code would behave, including `VectorIO`'s
insert protocol using `vector_db_id` in the request.
An alternative implementation would be to just delete the `vector_db_id`
parameter in `VectorDB` but the end result would still require users
having to write `vector_db_id = res.identifier` since
`VectorStores.create()` generates the ID for you.
So this approach felt the easiest way to migrate users towards
VectorStores (subsequent PRs will be added to trigger `files.create()`
and `vector_stores.files.create()`).
## Test Plan
Unit tests and integration tests have been added.
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
# What does this PR do?
Improved bedrock provider config to read from environment variables like
AWS_ACCESS_KEY_ID. Updated all
fields to use default_factory with lambda patterns like the nvidia
provider does.
Now the environment variables work as documented.
Closes#3305
## Test Plan
Ran the new bedrock config tests:
```bash
python -m pytest tests/unit/providers/inference/bedrock/test_config.py
-v
Verified existing provider tests still work:
python -m pytest tests/unit/providers/test_configs.py -v
What does this PR do?
Fixes error handling when MCP server connections fail. Instead of
returning generic 500 errors, now provides
descriptive error messages with proper HTTP status codes.
Closes#3107
Test Plan
Before fix:
curl -X GET
"http://localhost:8321/v1/tool-runtime/list-tools?tool_group_id=bad-mcp-server"
Returns: {"detail": "Internal server error: An unexpected error
occurred."} (500)
After fix:
curl -X GET
"http://localhost:8321/v1/tool-runtime/list-tools?tool_group_id=bad-mcp-server"
Returns: {"error": {"detail": "Failed to connect to MCP server at
http://localhost:9999/sse: Connection
refused"}} (502)
Tests:
- Added unit test for ConnectionError → 502 translation
- Manually tested with unreachable MCP servers (connection refused)
One needed to specify record-replay related environment variables for
running integration tests. We could not use defaults because integration
tests could be run against Ollama instances which could be running
different models. For example, text vs vision tests needed separate
instances of Ollama because a single instance typically cannot serve
both of these models if you assume the standard CI worker configuration
on Github. As a result, `client.list()` as returned by the Ollama client
would be different between these runs and we'd end up overwriting
responses.
This PR "solves" it by adding a small amount of complexity -- we store
model list responses specially, keyed by the hashes of the models they
return. At replay time, we merge all of them and pretend that we have
the union of all models available.
## Test Plan
Re-recorded all the tests using `scripts/integration-tests.sh
--inference-mode record`, including the vision tests.
