# What does this PR do?
This PR introduces APIs to retrieve past chat completion requests, which
will be used in the LS UI.
Our current `Telemetry` is ill-suited for this purpose as it's untyped
so we'd need to filter by obscure attribute names, making it brittle.
Since these APIs are 'provided by stack' and don't need to be
implemented by inference providers, we introduce a new InferenceProvider
class, containing the existing inference protocol, which is implemented
by inference providers.
The APIs are OpenAI-compliant, with an additional `input_messages`
field.
## Test Plan
This PR just adds the API and marks them provided_by_stack. S
tart stack server -> doesn't crash
note: the openai provider exposes the litellm specific model names to
the user. this change is compatible with that. the litellm names should
be deprecated.
# What does this PR do?
The goal of this PR is code base modernization.
Schema reflection code needed a minor adjustment to handle UnionTypes
and collections.abc.AsyncIterator. (Both are preferred for latest Python
releases.)
Note to reviewers: almost all changes here are automatically generated
by pyupgrade. Some additional unused imports were cleaned up. The only
change worth of note can be found under `docs/openapi_generator` and
`llama_stack/strong_typing/schema.py` where reflection code was updated
to deal with "newer" types.
Signed-off-by: Ihar Hrachyshka <ihar.hrachyshka@gmail.com>
# What does this PR do?
TLDR: Changes needed to get 100% passing tests for OpenAI API
verification tests when run against Llama Stack with the `together`,
`fireworks`, and `openai` providers. And `groq` is better than before,
at 88% passing.
This cleans up the OpenAI API support for image message types
(specifically `image_url` types) and handling of the `response_format`
chat completion parameter. Both of these required a few more Pydantic
model definitions in our Inference API, just to move from the
not-quite-right stubs I had in place to something fleshed out to match
the actual OpenAI API specs.
As part of testing this, I also found and fixed a bug in the litellm
implementation of openai_completion and openai_chat_completion, so the
providers based on those should actually be working now.
The method `prepare_openai_completion_params` in
`llama_stack/providers/utils/inference/openai_compat.py` was improved to
actually recursively clean up input parameters, including handling of
lists, dicts, and dumping of Pydantic models to dicts. These changes
were required to get to 100% passing tests on the OpenAI API
verification against the `openai` provider.
With the above, the together.ai provider was passing as well as it is
without Llama Stack. But, since we have Llama Stack in the middle, I
took the opportunity to clean up the together.ai provider so that it now
also passes the OpenAI API spec tests we have at 100%. That means
together.ai is now passing our verification test better when using an
OpenAI client talking to Llama Stack than it is when hitting together.ai
directly, without Llama Stack in the middle.
And, another round of work for Fireworks to improve translation of
incoming OpenAI chat completion requests to Llama Stack chat completion
requests gets the fireworks provider passing at 100%. The server-side
fireworks.ai tool calling support with OpenAI chat completions and Llama
4 models isn't great yet, but by pointing the OpenAI clients at Llama
Stack's API we can clean things up and get everything working as
expected for Llama 4 models.
## Test Plan
### OpenAI API Verification Tests
I ran the OpenAI API verification tests as below and 100% of the tests
passed.
First, start a Llama Stack server that runs the `openai` provider with
the `gpt-4o` and `gpt-4o-mini` models deployed. There's not a template
setup to do this out of the box, so I added a
`tests/verifications/openai-api-verification-run.yaml` to do this.
First, ensure you have the necessary API key environment variables set:
```
export TOGETHER_API_KEY="..."
export FIREWORKS_API_KEY="..."
export OPENAI_API_KEY="..."
```
Then, run a Llama Stack server that serves up all these providers:
```
llama stack run \
--image-type venv \
tests/verifications/openai-api-verification-run.yaml
```
Finally, generate a new verification report against all these providers,
both with and without the Llama Stack server in the middle.
```
python tests/verifications/generate_report.py \
--run-tests \
--provider \
together \
fireworks \
groq \
openai \
together-llama-stack \
fireworks-llama-stack \
groq-llama-stack \
openai-llama-stack
```
You'll see that most of the configurations with Llama Stack in the
middle now pass at 100%, even though some of them do not pass at 100%
when hitting the backend provider's API directly with an OpenAI client.
### OpenAI Completion Integration Tests with vLLM:
I also ran the smaller `test_openai_completion.py` test suite (that's
not yet merged with the verification tests) on multiple of the
providers, since I had to adjust the method signature of
openai_chat_completion a bit and thus had to touch lots of these
providers to match. Here's the tests I ran there, all passing:
```
VLLM_URL="http://localhost:8000/v1" INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct" llama stack build --template remote-vllm --image-type venv --run
```
in another terminal
```
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "meta-llama/Llama-3.2-3B-Instruct"
```
### OpenAI Completion Integration Tests with ollama
```
INFERENCE_MODEL="llama3.2:3b-instruct-q8_0" llama stack build --template ollama --image-type venv --run
```
in another terminal
```
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="llama3.2:3b-instruct-q8_0" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "llama3.2:3b-instruct-q8_0"
```
### OpenAI Completion Integration Tests with together.ai
```
INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct-Turbo" llama stack build --template together --image-type venv --run
```
in another terminal
```
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct-Turbo" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "meta-llama/Llama-3.2-3B-Instruct-Turbo"
```
### OpenAI Completion Integration Tests with fireworks.ai
```
INFERENCE_MODEL="meta-llama/Llama-3.1-8B-Instruct" llama stack build --template fireworks --image-type venv --run
```
in another terminal
```
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="meta-llama/Llama-3.1-8B-Instruct" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "meta-llama/Llama-3.1-8B-Instruct"
---------
Signed-off-by: Ben Browning <bbrownin@redhat.com>
# What does this PR do?
This PR adds two methods to the Inference API:
- `batch_completion`
- `batch_chat_completion`
The motivation is for evaluations targeting a local inference engine
(like meta-reference or vllm) where batch APIs provide for a substantial
amount of acceleration.
Why did I not add this to `Api.batch_inference` though? That just
resulted in a _lot_ more book-keeping given the structure of Llama
Stack. Had I done that, I would have needed to create a notion of a
"batch model" resource, setup routing based on that, etc. This does not
sound ideal.
So what's the future of the batch inference API? I am not sure. Maybe we
can keep it for true _asynchronous_ execution. So you can submit
requests, and it can return a Job instance, etc.
## Test Plan
Run meta-reference-gpu using:
```bash
export INFERENCE_MODEL=meta-llama/Llama-4-Scout-17B-16E-Instruct
export INFERENCE_CHECKPOINT_DIR=../checkpoints/Llama-4-Scout-17B-16E-Instruct-20250331210000
export MODEL_PARALLEL_SIZE=4
export MAX_BATCH_SIZE=32
export MAX_SEQ_LEN=6144
LLAMA_MODELS_DEBUG=1 llama stack run meta-reference-gpu
```
Then run the batch inference test case.
# What does this PR do?
This stubs in some OpenAI server-side compatibility with three new
endpoints:
/v1/openai/v1/models
/v1/openai/v1/completions
/v1/openai/v1/chat/completions
This gives common inference apps using OpenAI clients the ability to
talk to Llama Stack using an endpoint like
http://localhost:8321/v1/openai/v1 .
The two "v1" instances in there isn't awesome, but the thinking is that
Llama Stack's API is v1 and then our OpenAI compatibility layer is
compatible with OpenAI V1. And, some OpenAI clients implicitly assume
the URL ends with "v1", so this gives maximum compatibility.
The openai models endpoint is implemented in the routing layer, and just
returns all the models Llama Stack knows about.
The following providers should be working with the new OpenAI
completions and chat/completions API:
* remote::anthropic (untested)
* remote::cerebras-openai-compat (untested)
* remote::fireworks (tested)
* remote::fireworks-openai-compat (untested)
* remote::gemini (untested)
* remote::groq-openai-compat (untested)
* remote::nvidia (tested)
* remote::ollama (tested)
* remote::openai (untested)
* remote::passthrough (untested)
* remote::sambanova-openai-compat (untested)
* remote::together (tested)
* remote::together-openai-compat (untested)
* remote::vllm (tested)
The goal to support this for every inference provider - proxying
directly to the provider's OpenAI endpoint for OpenAI-compatible
providers. For providers that don't have an OpenAI-compatible API, we'll
add a mixin to translate incoming OpenAI requests to Llama Stack
inference requests and translate the Llama Stack inference responses to
OpenAI responses.
This is related to #1817 but is a bit larger in scope than just chat
completions, as I have real use-cases that need the older completions
API as well.
## Test Plan
### vLLM
```
VLLM_URL="http://localhost:8000/v1" INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct" llama stack build --template remote-vllm --image-type venv --run
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="meta-llama/Llama-3.2-3B-Instruct" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "meta-llama/Llama-3.2-3B-Instruct"
```
### ollama
```
INFERENCE_MODEL="llama3.2:3b-instruct-q8_0" llama stack build --template ollama --image-type venv --run
LLAMA_STACK_CONFIG=http://localhost:8321 INFERENCE_MODEL="llama3.2:3b-instruct-q8_0" python -m pytest -v tests/integration/inference/test_openai_completion.py --text-model "llama3.2:3b-instruct-q8_0"
```
## Documentation
Run a Llama Stack distribution that uses one of the providers mentioned
in the list above. Then, use your favorite OpenAI client to send
completion or chat completion requests with the base_url set to
http://localhost:8321/v1/openai/v1 . Replace "localhost:8321" with the
host and port of your Llama Stack server, if different.
---------
Signed-off-by: Ben Browning <bbrownin@redhat.com>
# Summary:
Includes fixes to get test_agents working with openAI model, e.g. tool
parsing and message conversion
# Test Plan:
```
LLAMA_STACK_CONFIG=dev pytest -s -v tests/integration/agents/test_agents.py --safety-shield meta-llama/Llama-Guard-3-8B --text-model openai/gpt-4o-mini
```
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with
[ReviewStack](https://reviewstack.dev/meta-llama/llama-stack/pull/1550).
* #1556
* __->__ #1550
# What does this PR do?
This commit introduces a new logging system that allows loggers to be
assigned
a category while retaining the logger name based on the file name. The
log
format includes both the logger name and the category, producing output
like:
```
INFO 2025-03-03 21:44:11,323 llama_stack.distribution.stack:103 [core]: Tool_groups: builtin::websearch served by
tavily-search
```
Key features include:
- Category-based logging: Loggers can be assigned a category (e.g.,
"core", "server") when programming. The logger can be loaded like
this: `logger = get_logger(name=__name__, category="server")`
- Environment variable control: Log levels can be configured
per-category using the
`LLAMA_STACK_LOGGING` environment variable. For example:
`LLAMA_STACK_LOGGING="server=DEBUG;core=debug"` enables DEBUG level for
the "server"
and "core" categories.
- `LLAMA_STACK_LOGGING="all=debug"` sets DEBUG level globally for all
categories and
third-party libraries.
This provides fine-grained control over logging levels while maintaining
a clean and
informative log format.
The formatter uses the rich library which provides nice colors better
stack traces like so:
```
ERROR 2025-03-03 21:49:37,124 asyncio:1758 [uncategorized]: unhandled exception during asyncio.run() shutdown
task: <Task finished name='Task-16' coro=<handle_signal.<locals>.shutdown() done, defined at
/Users/leseb/Documents/AI/llama-stack/llama_stack/distribution/server/server.py:146>
exception=UnboundLocalError("local variable 'loop' referenced before assignment")>
╭────────────────────────────────────── Traceback (most recent call last) ───────────────────────────────────────╮
│ /Users/leseb/Documents/AI/llama-stack/llama_stack/distribution/server/server.py:178 in shutdown │
│ │
│ 175 │ │ except asyncio.CancelledError: │
│ 176 │ │ │ pass │
│ 177 │ │ finally: │
│ ❱ 178 │ │ │ loop.stop() │
│ 179 │ │
│ 180 │ loop = asyncio.get_running_loop() │
│ 181 │ loop.create_task(shutdown()) │
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
UnboundLocalError: local variable 'loop' referenced before assignment
```
Co-authored-by: Ashwin Bharambe <@ashwinb>
Signed-off-by: Sébastien Han <seb@redhat.com>
[//]: # (If resolving an issue, uncomment and update the line below)
[//]: # (Closes #[issue-number])
## Test Plan
```
python -m llama_stack.distribution.server.server --yaml-config ./llama_stack/templates/ollama/run.yaml
INFO 2025-03-03 21:55:35,918 __main__:365 [server]: Using config file: llama_stack/templates/ollama/run.yaml
INFO 2025-03-03 21:55:35,925 __main__:378 [server]: Run configuration:
INFO 2025-03-03 21:55:35,928 __main__:380 [server]: apis:
- agents
```
[//]: # (## Documentation)
---------
Signed-off-by: Sébastien Han <seb@redhat.com>
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
# What does this PR do?
The commit addresses the Ruff warning B008 by refactoring the code to
avoid calling SamplingParams() directly in function argument defaults.
Instead, it either uses Field(default_factory=SamplingParams) for
Pydantic models or sets the default to None and instantiates
SamplingParams inside the function body when the argument is None.
Signed-off-by: Sébastien Han <seb@redhat.com>
A self-respecting server needs good observability which starts with
configurable logging. Llama Stack had little until now. This PR adds a
`logcat` facility towards that. Callsites look like:
```python
logcat.debug("inference", f"params to ollama: {params}")
```
- the first parameter is a category. there is a static list of
categories in `llama_stack/logcat.py`
- each category can be associated with a log-level which can be
configured via the `LLAMA_STACK_LOGGING` env var.
- a value `LLAMA_STACK_LOGGING=inference=debug;server=info"` does the
obvious thing. there is a special key called `all` which is an alias for
all categories
## Test Plan
Ran with `LLAMA_STACK_LOGGING="all=debug" llama stack run fireworks` and
saw the following:
Hit it with a client-sdk test case and saw this:
Groq has never supported raw completions anyhow. So this makes it easier
to switch it to LiteLLM. All our test suite passes.
I also updated all the openai-compat providers so they work with api
keys passed from headers. `provider_data`
## Test Plan
```bash
LLAMA_STACK_CONFIG=groq \
pytest -s -v tests/client-sdk/inference/test_text_inference.py \
--inference-model=groq/llama-3.3-70b-versatile --vision-inference-model=""
```
Also tested (openai, anthropic, gemini) providers. No regressions.
# What does this PR do?
This PR introduces more non-llama model support to llama stack.
Providers introduced: openai, anthropic and gemini. All of these
providers use essentially the same piece of code -- the implementation
works via the `litellm` library.
We will expose only specific models for providers we enable making sure
they all work well and pass tests. This setup (instead of automatically
enabling _all_ providers and models allowed by LiteLLM) ensures we can
also perform any needed prompt tuning on a per-model basis as needed
(just like we do it for llama models.)
## Test Plan
```bash
#!/bin/bash
args=("$@")
for model in openai/gpt-4o anthropic/claude-3-5-sonnet-latest gemini/gemini-1.5-flash; do
LLAMA_STACK_CONFIG=dev pytest -s -v tests/client-sdk/inference/test_text_inference.py \
--embedding-model=all-MiniLM-L6-v2 \
--vision-inference-model="" \
--inference-model=$model "${args[@]}"
done
```
