# What does this PR do?
Refactor setting default vector store provider and embedding model to
use an optional `vector_stores` config in the `StackRunConfig` and clean
up code to do so (had to add back in some pieces of VectorDB). Also
added remote Qdrant and Weaviate to starter distro (based on other PR
where inference providers were added for UX).
New config is simply (default for Starter distro):
```yaml
vector_stores:
default_provider_id: faiss
default_embedding_model:
provider_id: sentence-transformers
model_id: nomic-ai/nomic-embed-text-v1.5
```
## Test Plan
CI and Unit tests.
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
**This PR changes configurations in a backward incompatible way.**
Run configs today repeat full SQLite/Postgres snippets everywhere a
store is needed, which means duplicated credentials, extra connection
pools, and lots of drift between files. This PR introduces named storage
backends so the stack and providers can share a single catalog and
reference those backends by name.
## Key Changes
- Add `storage.backends` to `StackRunConfig`, register each KV/SQL
backend once at startup, and validate that references point to the right
family.
- Move server stores under `storage.stores` with lightweight references
(backend + namespace/table) instead of full configs.
- Update every provider/config/doc to use the new reference style;
docs/codegen now surface the simplified YAML.
## Migration
Before:
```yaml
metadata_store:
type: sqlite
db_path: ~/.llama/distributions/foo/registry.db
inference_store:
type: postgres
host: ${env.POSTGRES_HOST}
port: ${env.POSTGRES_PORT}
db: ${env.POSTGRES_DB}
user: ${env.POSTGRES_USER}
password: ${env.POSTGRES_PASSWORD}
conversations_store:
type: postgres
host: ${env.POSTGRES_HOST}
port: ${env.POSTGRES_PORT}
db: ${env.POSTGRES_DB}
user: ${env.POSTGRES_USER}
password: ${env.POSTGRES_PASSWORD}
```
After:
```yaml
storage:
backends:
kv_default:
type: kv_sqlite
db_path: ~/.llama/distributions/foo/kvstore.db
sql_default:
type: sql_postgres
host: ${env.POSTGRES_HOST}
port: ${env.POSTGRES_PORT}
db: ${env.POSTGRES_DB}
user: ${env.POSTGRES_USER}
password: ${env.POSTGRES_PASSWORD}
stores:
metadata:
backend: kv_default
namespace: registry
inference:
backend: sql_default
table_name: inference_store
max_write_queue_size: 10000
num_writers: 4
conversations:
backend: sql_default
table_name: openai_conversations
```
Provider configs follow the same pattern—for example, a Chroma vector
adapter switches from:
```yaml
providers:
vector_io:
- provider_id: chromadb
provider_type: remote::chromadb
config:
url: ${env.CHROMADB_URL}
kvstore:
type: sqlite
db_path: ~/.llama/distributions/foo/chroma.db
```
to:
```yaml
providers:
vector_io:
- provider_id: chromadb
provider_type: remote::chromadb
config:
url: ${env.CHROMADB_URL}
persistence:
backend: kv_default
namespace: vector_io::chroma_remote
```
Once the backends are declared, everything else just points at them, so
rotating credentials or swapping to Postgres happens in one place and
the stack reuses a single connection pool.
# Problem
The current inline provider appends the user provided instructions to
messages as a system prompt, but the returned response object does not
contain the instructions field (as specified in the OpenAI responses
spec).
# What does this PR do?
This pull request adds the instruction field to the response object
definition and updates the inline provider. It also ensures that
instructions from previous response is not carried over to the next
response (as specified in the openAI spec).
Closes #[3566](https://github.com/llamastack/llama-stack/issues/3566)
## Test Plan
- Tested manually for change in model response w.r.t supplied
instructions field.
- Added unit test to check that the instructions from previous response
is not carried over to the next response.
- Added integration tests to check instructions parameter in the
returned response object.
- Added new recordings for the integration tests.
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
# What does this PR do?
removes error:
ConnectionError: HTTPConnectionPool(host='localhost', port=4318): Max
retries exceeded with url: /v1/traces
(Caused by NewConnectionError('<urllib3.connection.HTTPConnection object
at 0x10fd98e60>: Failed to establish a
new connection: [Errno 61] Connection refused'))
## Test Plan
uv run llama stack run starter
curl http://localhost:8321/v1/models
observe no error in server logs
# What does this PR do?
<!-- 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] -->
- Fix examples in the NVIDIA inference documentation to align with
current API requirements.
## 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.* -->
N/A
# What does this PR do?
Adds a test and a standardized way to build future tests out for
telemetry in llama stack.
Contributes to https://github.com/llamastack/llama-stack/issues/3806
## Test Plan
This is the test plan 😎
# What does this PR do?
remove telemetry as a providable API from the codebase. This includes
removing it from generated distributions but also the provider registry,
the router, etc
since `setup_logger` is tied pretty strictly to `Api.telemetry` being in
impls we still need an "instantiated provider" in our implementations.
However it should not be auto-routed or provided. So in
validate_and_prepare_providers (called from resolve_impls) I made it so
that if run_config.telemetry.enabled, we set up the meta-reference
"provider" internally to be used so that log_event will work when
called.
This is the neatest way I think we can remove telemetry from the
provider configs but also not need to rip apart the whole "telemetry is
a provider" logic just yet, but we can do it internally later without
disrupting users.
so telemetry is removed from the registry such that if a user puts
`telemetry:` as an API in their build/run config it will err out, but
can still be used by us internally as we go through this transition.
relates to #3806
Signed-off-by: Charlie Doern <cdoern@redhat.com>
As indicated in the title. Our `starter` distribution enables all remote
providers _very intentionally_ because we believe it creates an easier,
more welcoming experience to new folks using the software. If we do
that, and then slam the logs with errors making them question their life
choices, it is not so good :)
Note that this fix is limited in scope. If you ever try to actually
instantiate the OpenAI client from a code path without an API key being
present, you deserve to fail hard.
## Test Plan
Run `llama stack run starter` with `OPENAI_API_KEY` set. No more wall of
text, just one message saying "listed 96 models".
a bunch of logger.info()s are good for server code to help debug in
production, but we don't want them killing our unit test output :)
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
**!!BREAKING CHANGE!!**
The lookup is also straightforward -- we always look for this identifier
and don't try to find a match for something without the provider_id
prefix.
Note that, this ideally means we need to update the `register_model()`
API also (we should kill "identifier" from there) but I am not doing
that as part of this PR.
## Test Plan
Existing unit tests
Wanted to re-enable Responses CI but it seems to hang for some reason
due to some interactions with conversations_store or responses_store.
## Test Plan
```
# library client
./scripts/integration-tests.sh --stack-config ci-tests --suite responses
# server
./scripts/integration-tests.sh --stack-config server:ci-tests --suite responses
```
# What does this PR do?
Have closed the previous PR due to merge conflicts with multiple PRs
Addressed all comments from
https://github.com/llamastack/llama-stack/pull/3768 (sorry for carrying
over to this one)
## Test Plan
Added UTs and integration tests
Handle a base case when no stored messages exist because no Response
call has been made.
## Test Plan
```
./scripts/integration-tests.sh --stack-config server:ci-tests \
--suite responses --inference-mode record-if-missing --pattern test_conversation_responses
```
Fixed KeyError when chunks don't have document_id in metadata or
chunk_metadata. Updated logging to safely extract document_id using
getattr and RAG memory to handle different document_id locations. Added
test for missing document_id scenarios.
Fixes issue #3494 where /v1/vector-io/insert would crash with KeyError.
Fixed KeyError when chunks don't have document_id in metadata or
chunk_metadata. Updated logging to safely extract document_id using
getattr and RAG memory to handle different document_id locations. Added
test for missing document_id scenarios.
# What does this PR do?
Fixes a KeyError crash in `/v1/vector-io/insert` when chunks are missing
`document_id` fields. The API
was failing even though `document_id` is optional according to the
schema.
Closes#3494
## Test Plan
**Before fix:**
- POST to `/v1/vector-io/insert` with chunks → 500 KeyError
- Happened regardless of where `document_id` was placed
**After fix:**
- Same request works fine → 200 OK
- Tested with Postman using FAISS backend
- Added unit test covering missing `document_id` scenarios
This PR updates the Conversation item related types and improves a
couple critical parts of the implemenation:
- it creates a streaming output item for the final assistant message
output by
the model. until now we only added content parts and included that
message in the final response.
- rewrites the conversation update code completely to account for items
other than messages (tool calls, outputs, etc.)
## Test Plan
Used the test script from
https://github.com/llamastack/llama-stack-client-python/pull/281 for
this
```
TEST_API_BASE_URL=http://localhost:8321/v1 \
pytest tests/integration/test_agent_turn_step_events.py::test_client_side_function_tool -xvs
```
# Add support for Google Gemini `gemini-embedding-001` embedding model
and correctly registers model type
MR message created with the assistance of Claude-4.5-sonnet
This resolves https://github.com/llamastack/llama-stack/issues/3755
## What does this PR do?
This PR adds support for the `gemini-embedding-001` Google embedding
model to the llama-stack Gemini provider. This model provides
high-dimensional embeddings (3072 dimensions) compared to the existing
`text-embedding-004` model (768 dimensions). Old embeddings models (such
as text-embedding-004) will be deprecated soon according to Google
([Link](https://developers.googleblog.com/en/gemini-embedding-available-gemini-api/))
## Problem
The Gemini provider only supported the `text-embedding-004` embedding
model. The newer `gemini-embedding-001` model, which provides
higher-dimensional embeddings for improved semantic representation, was
not available through llama-stack.
## Solution
This PR consists of three commits that implement, fix the model
registration, and enable embedding generation:
### Commit 1: Initial addition of gemini-embedding-001
Added metadata for `gemini-embedding-001` to the
`embedding_model_metadata` dictionary:
```python
embedding_model_metadata: dict[str, dict[str, int]] = {
"text-embedding-004": {"embedding_dimension": 768, "context_length": 2048},
"gemini-embedding-001": {"embedding_dimension": 3072, "context_length": 2048}, # NEW
}
```
**Issue discovered:** The model was not being registered correctly
because the dictionary keys didn't match the model IDs returned by
Gemini's API.
### Commit 2: Fix model ID matching with `models/` prefix
Updated both dictionary keys to include the `models/` prefix to match
Gemini's OpenAI-compatible API response format:
```python
embedding_model_metadata: dict[str, dict[str, int]] = {
"models/text-embedding-004": {"embedding_dimension": 768, "context_length": 2048}, # UPDATED
"models/gemini-embedding-001": {"embedding_dimension": 3072, "context_length": 2048}, # UPDATED
}
```
**Root cause:** Gemini's OpenAI-compatible API returns model IDs with
the `models/` prefix (e.g., `models/text-embedding-004`). The
`OpenAIMixin.list_models()` method directly matches these IDs against
the `embedding_model_metadata` dictionary keys. Without the prefix, the
models were being registered as LLMs instead of embedding models.
### Commit 3: Fix embedding generation for providers without usage stats
Fixed a bug in `OpenAIMixin.openai_embeddings()` that prevented
embedding generation for providers (like Gemini) that don't return usage
statistics:
```python
# Before (Line 351-354):
usage = OpenAIEmbeddingUsage(
prompt_tokens=response.usage.prompt_tokens, # ← Crashed with AttributeError
total_tokens=response.usage.total_tokens,
)
# After (Lines 351-362):
if response.usage:
usage = OpenAIEmbeddingUsage(
prompt_tokens=response.usage.prompt_tokens,
total_tokens=response.usage.total_tokens,
)
else:
usage = OpenAIEmbeddingUsage(
prompt_tokens=0, # Default when not provided
total_tokens=0, # Default when not provided
)
```
**Impact:** This fix enables embedding generation for **all** Gemini
embedding models, not just the newly added one.
## Changes
### Modified Files
**`llama_stack/providers/remote/inference/gemini/gemini.py`**
- Line 17: Updated `text-embedding-004` key to
`models/text-embedding-004`
- Line 18: Added `models/gemini-embedding-001` with correct metadata
**`llama_stack/providers/utils/inference/openai_mixin.py`**
- Lines 351-362: Added null check for `response.usage` to handle
providers without usage statistics
## Key Technical Details
### Model ID Matching Flow
1. `list_provider_model_ids()` calls Gemini's `/v1/models` endpoint
2. API returns model IDs like: `models/text-embedding-004`,
`models/gemini-embedding-001`
3. `OpenAIMixin.list_models()` (line 410) checks: `if metadata :=
self.embedding_model_metadata.get(provider_model_id)`
4. If matched, registers as `model_type: "embedding"` with metadata;
otherwise registers as `model_type: "llm"`
### Why Both Keys Needed the Prefix
The `text-embedding-004` model was already working because there was
likely separate configuration or manual registration handling it. For
auto-discovery to work correctly for **both** models, both keys must
match the API's model ID format exactly.
## How to test this PR
Verified the changes by:
1. **Model Auto-Discovery**: Started llama-stack server and confirmed
models are auto-discovered from Gemini API
2. **Model Registration**: Confirmed both embedding models are correctly
registered and visible
```bash
curl http://localhost:8325/v1/models | jq '.data[] | select(.provider_id == "gemini" and .model_type == "embedding")'
```
**Results:**
- ✅ `gemini/models/text-embedding-004` - 768 dimensions - `model_type:
"embedding"`
- ✅ `gemini/models/gemini-embedding-001` - 3072 dimensions -
`model_type: "embedding"`
3. **Before Fix (Commit 1)**: Models appeared as `model_type: "llm"`
without embedding metadata
4. **After Fix (Commit 2)**: Models correctly identified as `model_type:
"embedding"` with proper metadata
5. **Generate Embeddings**: Verified embedding generation works
```bash
curl -X POST http://localhost:8325/v1/embeddings \
-H "Content-Type: application/json" \
-d '{"model": "gemini/models/gemini-embedding-001", "input": "test"}' | \
jq '.data[0].embedding | length'
```
# What does this PR do?
Enables automatic embedding model detection for vector stores and by
using a `default_configured` boolean that can be defined in the
`run.yaml`.
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
## Test Plan
- Unit tests
- Integration tests
- Simple example below:
Spin up the stack:
```bash
uv run llama stack build --distro starter --image-type venv --run
```
Then test with OpenAI's client:
```python
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8321/v1/", api_key="none")
vs = client.vector_stores.create()
```
Previously you needed:
```python
vs = client.vector_stores.create(
extra_body={
"embedding_model": "sentence-transformers/all-MiniLM-L6-v2",
"embedding_dimension": 384,
}
)
```
The `extra_body` is now unnecessary.
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
Previously, the NVIDIA inference provider implemented a custom
`openai_embeddings` method with a hardcoded `input_type="query"`
parameter, which is required by NVIDIA asymmetric embedding
models([https://github.com/llamastack/llama-stack/pull/3205](https://github.com/llamastack/llama-stack/pull/3205)).
Recently `extra_body` parameter is added to the embeddings API
([https://github.com/llamastack/llama-stack/pull/3794](https://github.com/llamastack/llama-stack/pull/3794)).
So, this PR updates the NVIDIA inference provider to use the base
`OpenAIMixin.openai_embeddings` method instead and pass the `input_type`
through the `extra_body` parameter for asymmetric embedding models.
<!-- 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.* -->
Run the following command for the ```embedding_model```:
```nvidia/llama-3.2-nv-embedqa-1b-v2```, ```nvidia/nv-embedqa-e5-v5```,
```nvidia/nv-embedqa-mistral-7b-v2```, and
```snowflake/arctic-embed-l```.
```
pytest -s -v tests/integration/inference/test_openai_embeddings.py --stack-config="inference=nvidia" --embedding-model={embedding_model} --env NVIDIA_API_KEY={nvidia_api_key} --env NVIDIA_BASE_URL="https://integrate.api.nvidia.com" --inference-mode=record
```
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
The purpose of this PR is to replace the Llama Stack's default embedding
model by nomic-embed-text-v1.5.
These are the key reasons why Llama Stack community decided to switch
from all-MiniLM-L6-v2 to nomic-embed-text-v1.5:
1. The training data for
[all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2#training-data)
includes a lot of data sets with various licensing terms, so it is
tricky to know when/whether it is appropriate to use this model for
commercial applications.
2. The model is not particularly competitive on major benchmarks. For
example, if you look at the [MTEB
Leaderboard](https://huggingface.co/spaces/mteb/leaderboard) and click
on Miscellaneous/BEIR to see English information retrieval accuracy, you
see that the top of the leaderboard is dominated by enormous models but
also that there are many, many models of relatively modest size whith
much higher Retrieval scores. If you want to look closely at the data, I
recommend clicking "Download Table" because it is easier to browse that
way.
More discussion info can be founded
[here](https://github.com/llamastack/llama-stack/issues/2418)
<!-- If resolving an issue, uncomment and update the line below -->
<!-- Closes #[issue-number] -->
Closes#2418
## 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.* -->
1. Run `./scripts/unit-tests.sh`
2. Integration tests via CI wokrflow
---------
Signed-off-by: Sébastien Han <seb@redhat.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Francisco Arceo <arceofrancisco@gmail.com>
Co-authored-by: Sébastien Han <seb@redhat.com>
# What does this PR do?
This PR fixes issues with the WatsonX provider so it works correctly
with LiteLLM.
The main problem was that WatsonX requests failed because the provider
data validator didn’t properly handle the API key and project ID. This
was fixed by updating the WatsonXProviderDataValidator and ensuring the
provider data is loaded correctly.
The openai_chat_completion method was also updated to match the behavior
of other providers while adding WatsonX-specific fields like project_id.
It still calls await super().openai_chat_completion.__func__(self,
params) to keep the existing setup and tracing logic.
After these changes, WatsonX requests now run correctly.
## Test Plan
The changes were tested by running chat completion requests and
confirming that credentials and project parameters are passed correctly.
I have tested with my WatsonX credentials, by using the cli with `uv run
llama-stack-client inference chat-completion --session`
---------
Signed-off-by: Sébastien Han <seb@redhat.com>
Co-authored-by: Sébastien Han <seb@redhat.com>
# What does this PR do?
2 main changes:
1. Remove `provider_id` requirement in call to vector stores and
2. Removes "register first embedding model" logic
- Now forces embedding model id as required on Vector Store creation
Simplifies the UX for OpenAI to:
```python
vs = client.vector_stores.create(
name="my_citations_db",
extra_body={
"embedding_model": "ollama/nomic-embed-text:latest",
}
)
```
<!-- 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.* -->
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Applies the same pattern from
https://github.com/llamastack/llama-stack/pull/3777 to embeddings and
vector_stores.create() endpoints.
This should _not_ be a breaking change since (a) our tests were already
using the `extra_body` parameter when passing in to the backend (b) but
the backend probably wasn't extracting the parameters correctly. This PR
will fix that.
Updated APIs: `openai_embeddings(), openai_create_vector_store(),
openai_create_vector_store_file_batch()`
Implements missing streaming events from OpenAI Responses API spec:
- reasoning text/summary events for o1/o3 models,
- refusal events for safety moderation
- annotation events for citations,
- and file search streaming events.
Added optional reasoning_content field to chat completion chunks to
support non-standard provider extensions.
**NOTE:** OpenAI does _not_ fill reasoning_content when users use the
chat_completion APIs. This means there is no way for us to implement
Responses (with reasoning) by using OpenAI chat completions! We'd need
to transparently punt to OpenAI's responses endpoints if we wish to do
that. For others though (vLLM, etc.) we can use it.
## Test Plan
File search streaming test passes:
```
./scripts/integration-tests.sh --stack-config server:ci-tests \
--suite responses --setup gpt --inference-mode replay --pattern test_response_file_search_streaming_events
```
Need more complex setup and validation for reasoning tests (need a vLLM
powered OSS model maybe gpt-oss which can return reasoning_content). I
will do that in a followup PR.
# What does this PR do?
Removes VectorDBs from API surface and our tests.
Moves tests to Vector Stores.
<!-- 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.* -->
---------
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
# What does this PR do?
Allows passing through extra_body parameters to inference providers.
With this, we removed the 2 vllm-specific parameters from completions
API into `extra_body`.
Before/After
<img width="1883" height="324" alt="image"
src="https://github.com/user-attachments/assets/acb27c08-c748-46c9-b1da-0de64e9908a1"
/>
closes#2720
## Test Plan
CI and added new test
```
❯ uv run pytest -s -v tests/integration/ --stack-config=server:starter --inference-mode=record -k 'not( builtin_tool or safety_with_image or code_interpreter or test_rag ) and test_openai_completion_guided_choice' --setup=vllm --suite=base --color=yes
Uninstalled 3 packages in 125ms
Installed 3 packages in 19ms
INFO 2025-10-10 14:29:54,317 tests.integration.conftest:118 tests: Applying setup 'vllm' for suite base
INFO 2025-10-10 14:29:54,331 tests.integration.conftest:47 tests: Test stack config type: server
(stack_config=server:starter)
============================================================================================================== test session starts ==============================================================================================================
platform darwin -- Python 3.12.11, pytest-8.4.2, pluggy-1.6.0 -- /Users/erichuang/projects/llama-stack-1/.venv/bin/python
cachedir: .pytest_cache
metadata: {'Python': '3.12.11', '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/erichuang/projects/llama-stack-1
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 285 items / 284 deselected / 1 selected
tests/integration/inference/test_openai_completion.py::test_openai_completion_guided_choice[txt=vllm/Qwen/Qwen3-0.6B]
instantiating llama_stack_client
Starting llama stack server with config 'starter' on port 8321...
Waiting for server at http://localhost:8321... (0.0s elapsed)
Waiting for server at http://localhost:8321... (0.5s elapsed)
Waiting for server at http://localhost:8321... (5.1s elapsed)
Waiting for server at http://localhost:8321... (5.6s elapsed)
Waiting for server at http://localhost:8321... (10.1s elapsed)
Waiting for server at http://localhost:8321... (10.6s elapsed)
Server is ready at http://localhost:8321
llama_stack_client instantiated in 11.773s
PASSEDTerminating llama stack server process...
Terminating process 98444 and its group...
Server process and children terminated gracefully
============================================================================================================= slowest 10 durations ==============================================================================================================
11.88s setup tests/integration/inference/test_openai_completion.py::test_openai_completion_guided_choice[txt=vllm/Qwen/Qwen3-0.6B]
3.02s call tests/integration/inference/test_openai_completion.py::test_openai_completion_guided_choice[txt=vllm/Qwen/Qwen3-0.6B]
0.01s teardown tests/integration/inference/test_openai_completion.py::test_openai_completion_guided_choice[txt=vllm/Qwen/Qwen3-0.6B]
================================================================================================ 1 passed, 284 deselected, 3 warnings in 16.21s =================================================================================================
```
# What does this PR do?
Converts openai(_chat)_completions params to pydantic BaseModel to
reduce code duplication across all providers.
## Test Plan
CI
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with
[ReviewStack](https://reviewstack.dev/llamastack/llama-stack/pull/3761).
* #3777
* __->__ #3761
Implementats usage accumulation to StreamingResponseOrchestrator.
The most important part was to pass `stream_options = { "include_usage":
true }` to the chat_completion call. This means I will have to record
all responses tests again because request hash will change :)
Test changes:
- Add usage assertions to streaming and non-streaming tests
- Update test recordings with actual usage data from OpenAI
# What does this PR do?
This PR checks whether, if a previous response is linked, there are
mcp_list_tools objects that can be reused instead of listing the tools
explicitly every time.
Closes#3106
## Test Plan
Tested manually.
Added unit tests to cover new behaviour.
---------
Signed-off-by: Gordon Sim <gsim@redhat.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
# What does this PR do?
use SecretStr for OpenAIMixin providers
- RemoteInferenceProviderConfig now has auth_credential: SecretStr
- the default alias is api_key (most common name)
- some providers override to use api_token (RunPod, vLLM, Databricks)
- some providers exclude it (Ollama, TGI, Vertex AI)
addresses #3517
## Test Plan
ci w/ new tests
# What does this PR do?
<!-- Provide a short summary of what this PR does and why. Link to
relevant issues if applicable. -->
Allows model check to fail gracefully instead of crashing on startup.
<!-- 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.* -->
set VLLM_URL to your VLLM server
```
(base) akram@Mac llama-stack % LAMA_STACK_LOGGING="all=debug" VLLM_ENABLE_MODEL_DISCOVERY=false MILVUS_DB_PATH=./milvus.db INFERENCE_MODEL=vllm uv run --with llama-stack llama stack build --distro starter --image-type venv --run
```
```
INFO 2025-10-08 20:11:24,637 llama_stack.providers.utils.inference.inference_store:74 inference: Write queue disabled for SQLite to avoid concurrency issues
INFO 2025-10-08 20:11:24,866 llama_stack.providers.utils.responses.responses_store:96 openai_responses: Write queue disabled for SQLite to avoid concurrency issues
ERROR 2025-10-08 20:11:26,160 llama_stack.providers.utils.inference.openai_mixin:439 providers::utils: VLLMInferenceAdapter.list_provider_model_ids() failed with: <a
href="https://oauth.akram.a1ey.p3.openshiftapps.com:443/oauth/authorize?approval_prompt=force&client_id=system%3Aserviceaccount%3Arhoai-30-genai%3Adefault&redirect_uri=ht
tps%3A%2F%2Fvllm-rhoai-30-genai.apps.rosa.akram.a1ey.p3.openshiftapps.com%2Foauth%2Fcallback&response_type=code&scope=user%3Ainfo+user%3Acheck-access&state=9fba207425
5851c718aca717a5887d76%3A%2Fmodels">Found</a>.
[...]
INFO 2025-10-08 20:11:26,295 uvicorn.error:84 uncategorized: Started server process [83144]
INFO 2025-10-08 20:11:26,296 uvicorn.error:48 uncategorized: Waiting for application startup.
INFO 2025-10-08 20:11:26,297 llama_stack.core.server.server:170 core::server: Starting up
INFO 2025-10-08 20:11:26,297 llama_stack.core.stack:399 core: starting registry refresh task
INFO 2025-10-08 20:11:26,311 uvicorn.error:62 uncategorized: Application startup complete.
INFO 2025-10-08 20:11:26,312 uvicorn.error:216 uncategorized: Uvicorn running on http://['::', '0.0.0.0']:8321 (Press CTRL+C to quit)
ERROR 2025-10-08 20:11:26,791 llama_stack.providers.utils.inference.openai_mixin:439 providers::utils: VLLMInferenceAdapter.list_provider_model_ids() failed with: <a
href="https://oauth.akram.a1ey.p3.openshiftapps.com:443/oauth/authorize?approval_prompt=force&client_id=system%3Aserviceaccount%3Arhoai-30-genai%3Adefault&redirect_uri=ht
tps%3A%2F%2Fvllm-rhoai-30-genai.apps.rosa.akram.a1ey.p3.openshiftapps.com%2Foauth%2Fcallback&response_type=code&scope=user%3Ainfo+user%3Acheck-access&state=8ef0cba3e1
71a4f8b04cb445cfb91a4c%3A%2Fmodels">Found</a>.
```
## Summary
After removing model management CLI in #3700, this PR updates remaining
references to the old `llama download` command to use `huggingface-cli
download` instead.
## Changes
- Updated error messages in `meta_reference/common.py` to recommend
`huggingface-cli download`
- Updated error messages in
`torchtune/recipes/lora_finetuning_single_device.py` to use
`huggingface-cli download`
- Updated post-training notebook to use `huggingface-cli download`
instead of `llama download`
- Fixed typo: "you model" -> "your model"
## Test Plan
- Verified error messages provide correct guidance for users
- Checked that notebook instructions are up-to-date with current tooling
Renames `inference_recorder.py` to `api_recorder.py` and extends it to
support recording/replaying tool invocations in addition to inference
calls.
This allows us to record web-search, etc. tool calls and thereafter
apply recordings for `tests/integration/responses`
## Test Plan
```
export OPENAI_API_KEY=...
export TAVILY_SEARCH_API_KEY=...
./scripts/integration-tests.sh --stack-config ci-tests \
--suite responses --inference-mode record-if-missing
```
# What does this PR do?
Adds traces around tool execution and mcp tool listing for better
observability.
Closes#3108
## Test Plan
Manually examined traces in jaeger to verify the added information was
available.
Signed-off-by: Gordon Sim <gsim@redhat.com>
# What does this PR do?
It prevents a tool call message being added to the chat completions
message without a corresponding tool call result, which is needed in the
case that an approval is required first or if the approval request is
denied. In both these cases the tool call messages is popped of the next
turn messages.
Closes#3728
## Test Plan
Ran the integration tests
Manual check of both approval and denial against gpt-4o
Signed-off-by: Gordon Sim <gsim@redhat.com>
# What does this PR do?
- The watsonx.ai provider now uses the LiteLLM mixin instead of using
IBM's library, which does not seem to be working (see #3165 for
context).
- The watsonx.ai provider now lists all the models available by calling
the watsonx.ai server instead of having a hard coded list of known
models. (That list gets out of date quickly)
- An edge case in
[llama_stack/core/routers/inference.py](https://github.com/llamastack/llama-stack/pull/3674/files#diff-a34bc966ed9befd9f13d4883c23705dff49be0ad6211c850438cdda6113f3455)
is addressed that was causing my manual tests to fail.
- Fixes `b64_encode_openai_embeddings_response` which was trying to
enumerate over a dictionary and then reference elements of the
dictionary using .field instead of ["field"]. That method is called by
the LiteLLM mixin for embedding models, so it is needed to get the
watsonx.ai embedding models to work.
- A unit test along the lines of the one in #3348 is added. A more
comprehensive plan for automatically testing the end-to-end
functionality for inference providers would be a good idea, but is out
of scope for this PR.
- Updates to the watsonx distribution. Some were in response to the
switch to LiteLLM (e.g., updating the Python packages needed). Others
seem to be things that were already broken that I found along the way
(e.g., a reference to a watsonx specific doc template that doesn't seem
to exist).
Closes#3165
Also it is related to a line-item in #3387 but doesn't really address
that goal (because it uses the LiteLLM mixin, not the OpenAI one). I
tried the OpenAI one and it doesn't work with watsonx.ai, presumably
because the watsonx.ai service is not OpenAI compatible. It works with
LiteLLM because LiteLLM has a provider implementation for watsonx.ai.
## Test Plan
The test script below goes back and forth between the OpenAI and watsonx
providers. The idea is that the OpenAI provider shows how it should work
and then the watsonx provider output shows that it is also working with
watsonx. Note that the result from the MCP test is not as good (the
Llama 3.3 70b model does not choose tools as wisely as gpt-4o), but it
is still working and providing a valid response. For more details on
setup and the MCP server being used for testing, see [the AI Alliance
sample
notebook](https://github.com/The-AI-Alliance/llama-stack-examples/blob/main/notebooks/01-responses/)
that these examples are drawn from.
```python
#!/usr/bin/env python3
import json
from llama_stack_client import LlamaStackClient
from litellm import completion
import http.client
def print_response(response):
"""Print response in a nicely formatted way"""
print(f"ID: {response.id}")
print(f"Status: {response.status}")
print(f"Model: {response.model}")
print(f"Created at: {response.created_at}")
print(f"Output items: {len(response.output)}")
for i, output_item in enumerate(response.output):
if len(response.output) > 1:
print(f"\n--- Output Item {i+1} ---")
print(f"Output type: {output_item.type}")
if output_item.type in ("text", "message"):
print(f"Response content: {output_item.content[0].text}")
elif output_item.type == "file_search_call":
print(f" Tool Call ID: {output_item.id}")
print(f" Tool Status: {output_item.status}")
# 'queries' is a list, so we join it for clean printing
print(f" Queries: {', '.join(output_item.queries)}")
# Display results if they exist, otherwise note they are empty
print(f" Results: {output_item.results if output_item.results else 'None'}")
elif output_item.type == "mcp_list_tools":
print_mcp_list_tools(output_item)
elif output_item.type == "mcp_call":
print_mcp_call(output_item)
else:
print(f"Response content: {output_item.content}")
def print_mcp_call(mcp_call):
"""Print MCP call in a nicely formatted way"""
print(f"\n🛠️ MCP Tool Call: {mcp_call.name}")
print(f" Server: {mcp_call.server_label}")
print(f" ID: {mcp_call.id}")
print(f" Arguments: {mcp_call.arguments}")
if mcp_call.error:
print("Error: {mcp_call.error}")
elif mcp_call.output:
print("Output:")
# Try to format JSON output nicely
try:
parsed_output = json.loads(mcp_call.output)
print(json.dumps(parsed_output, indent=4))
except:
# If not valid JSON, print as-is
print(f" {mcp_call.output}")
else:
print(" ⏳ No output yet")
def print_mcp_list_tools(mcp_list_tools):
"""Print MCP list tools in a nicely formatted way"""
print(f"\n🔧 MCP Server: {mcp_list_tools.server_label}")
print(f" ID: {mcp_list_tools.id}")
print(f" Available Tools: {len(mcp_list_tools.tools)}")
print("=" * 80)
for i, tool in enumerate(mcp_list_tools.tools, 1):
print(f"\n{i}. {tool.name}")
print(f" Description: {tool.description}")
# Parse and display input schema
schema = tool.input_schema
if schema and 'properties' in schema:
properties = schema['properties']
required = schema.get('required', [])
print(" Parameters:")
for param_name, param_info in properties.items():
param_type = param_info.get('type', 'unknown')
param_desc = param_info.get('description', 'No description')
required_marker = " (required)" if param_name in required else " (optional)"
print(f" • {param_name} ({param_type}){required_marker}")
if param_desc:
print(f" {param_desc}")
if i < len(mcp_list_tools.tools):
print("-" * 40)
def main():
"""Main function to run all the tests"""
# Configuration
LLAMA_STACK_URL = "http://localhost:8321/"
LLAMA_STACK_MODEL_IDS = [
"openai/gpt-3.5-turbo",
"openai/gpt-4o",
"llama-openai-compat/Llama-3.3-70B-Instruct",
"watsonx/meta-llama/llama-3-3-70b-instruct"
]
# Using gpt-4o for this demo, but feel free to try one of the others or add more to run.yaml.
OPENAI_MODEL_ID = LLAMA_STACK_MODEL_IDS[1]
WATSONX_MODEL_ID = LLAMA_STACK_MODEL_IDS[-1]
NPS_MCP_URL = "http://localhost:3005/sse/"
print("=== Llama Stack Testing Script ===")
print(f"Using OpenAI model: {OPENAI_MODEL_ID}")
print(f"Using WatsonX model: {WATSONX_MODEL_ID}")
print(f"MCP URL: {NPS_MCP_URL}")
print()
# Initialize client
print("Initializing LlamaStackClient...")
client = LlamaStackClient(base_url="http://localhost:8321")
# Test 1: List models
print("\n=== Test 1: List Models ===")
try:
models = client.models.list()
print(f"Found {len(models)} models")
except Exception as e:
print(f"Error listing models: {e}")
raise e
# Test 2: Basic chat completion with OpenAI
print("\n=== Test 2: Basic Chat Completion (OpenAI) ===")
try:
chat_completion_response = client.chat.completions.create(
model=OPENAI_MODEL_ID,
messages=[{"role": "user", "content": "What is the capital of France?"}]
)
print("OpenAI Response:")
for chunk in chat_completion_response.choices[0].message.content:
print(chunk, end="", flush=True)
print()
except Exception as e:
print(f"Error with OpenAI chat completion: {e}")
raise e
# Test 3: Basic chat completion with WatsonX
print("\n=== Test 3: Basic Chat Completion (WatsonX) ===")
try:
chat_completion_response_wxai = client.chat.completions.create(
model=WATSONX_MODEL_ID,
messages=[{"role": "user", "content": "What is the capital of France?"}],
)
print("WatsonX Response:")
for chunk in chat_completion_response_wxai.choices[0].message.content:
print(chunk, end="", flush=True)
print()
except Exception as e:
print(f"Error with WatsonX chat completion: {e}")
raise e
# Test 4: Tool calling with OpenAI
print("\n=== Test 4: Tool Calling (OpenAI) ===")
tools = [
{
"type": "function",
"function": {
"name": "get_current_weather",
"description": "Get the current weather for a specific location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g., San Francisco, CA",
},
"unit": {
"type": "string",
"enum": ["celsius", "fahrenheit"]
},
},
"required": ["location"],
},
},
}
]
messages = [
{"role": "user", "content": "What's the weather like in Boston, MA?"}
]
try:
print("--- Initial API Call ---")
response = client.chat.completions.create(
model=OPENAI_MODEL_ID,
messages=messages,
tools=tools,
tool_choice="auto", # "auto" is the default
)
print("OpenAI tool calling response received")
except Exception as e:
print(f"Error with OpenAI tool calling: {e}")
raise e
# Test 5: Tool calling with WatsonX
print("\n=== Test 5: Tool Calling (WatsonX) ===")
try:
wxai_response = client.chat.completions.create(
model=WATSONX_MODEL_ID,
messages=messages,
tools=tools,
tool_choice="auto", # "auto" is the default
)
print("WatsonX tool calling response received")
except Exception as e:
print(f"Error with WatsonX tool calling: {e}")
raise e
# Test 6: Streaming with WatsonX
print("\n=== Test 6: Streaming Response (WatsonX) ===")
try:
chat_completion_response_wxai_stream = client.chat.completions.create(
model=WATSONX_MODEL_ID,
messages=[{"role": "user", "content": "What is the capital of France?"}],
stream=True
)
print("Model response: ", end="")
for chunk in chat_completion_response_wxai_stream:
# Each 'chunk' is a ChatCompletionChunk object.
# We want the content from the 'delta' attribute.
if hasattr(chunk, 'choices') and chunk.choices is not None:
content = chunk.choices[0].delta.content
# The first few chunks might have None content, so we check for it.
if content is not None:
print(content, end="", flush=True)
print()
except Exception as e:
print(f"Error with streaming: {e}")
raise e
# Test 7: MCP with OpenAI
print("\n=== Test 7: MCP Integration (OpenAI) ===")
try:
mcp_llama_stack_client_response = client.responses.create(
model=OPENAI_MODEL_ID,
input="Tell me about some parks in Rhode Island, and let me know if there are any upcoming events at them.",
tools=[
{
"type": "mcp",
"server_url": NPS_MCP_URL,
"server_label": "National Parks Service tools",
"allowed_tools": ["search_parks", "get_park_events"],
}
]
)
print_response(mcp_llama_stack_client_response)
except Exception as e:
print(f"Error with MCP (OpenAI): {e}")
raise e
# Test 8: MCP with WatsonX
print("\n=== Test 8: MCP Integration (WatsonX) ===")
try:
mcp_llama_stack_client_response = client.responses.create(
model=WATSONX_MODEL_ID,
input="What is the capital of France?"
)
print_response(mcp_llama_stack_client_response)
except Exception as e:
print(f"Error with MCP (WatsonX): {e}")
raise e
# Test 9: MCP with Llama 3.3
print("\n=== Test 9: MCP Integration (Llama 3.3) ===")
try:
mcp_llama_stack_client_response = client.responses.create(
model=WATSONX_MODEL_ID,
input="Tell me about some parks in Rhode Island, and let me know if there are any upcoming events at them.",
tools=[
{
"type": "mcp",
"server_url": NPS_MCP_URL,
"server_label": "National Parks Service tools",
"allowed_tools": ["search_parks", "get_park_events"],
}
]
)
print_response(mcp_llama_stack_client_response)
except Exception as e:
print(f"Error with MCP (Llama 3.3): {e}")
raise e
# Test 10: Embeddings
print("\n=== Test 10: Embeddings ===")
try:
conn = http.client.HTTPConnection("localhost:8321")
payload = json.dumps({
"model": "watsonx/ibm/granite-embedding-278m-multilingual",
"input": "Hello, world!",
})
headers = {
'Content-Type': 'application/json',
'Accept': 'application/json'
}
conn.request("POST", "/v1/openai/v1/embeddings", payload, headers)
res = conn.getresponse()
data = res.read()
print(data.decode("utf-8"))
except Exception as e:
print(f"Error with Embeddings: {e}")
raise e
print("\n=== Testing Complete ===")
if __name__ == "__main__":
main()
```
---------
Signed-off-by: Bill Murdock <bmurdock@redhat.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
# What does this PR do?
Removing some dead code, found by vulture and checked by claude that
there are no references or imports for these
## Test Plan
CI
