# What does this PR do?
- braintrust scoring provider requires OPENAI_API_KEY env variable to be
set
- move this to be able to be set as request headers (e.g. like together
/ fireworks api keys)
- fixes pytest with agents dependency
## Test Plan
**E2E**
```
llama stack run
```
```yaml
scoring:
- provider_id: braintrust-0
provider_type: inline::braintrust
config: {}
```
**Client**
```python
self.client = LlamaStackClient(
base_url=os.environ.get("LLAMA_STACK_ENDPOINT", "http://localhost:5000"),
provider_data={
"openai_api_key": os.environ.get("OPENAI_API_KEY", ""),
},
)
```
- run `llama-stack-client eval run_scoring`
**Unit Test**
```
pytest -v -s -m meta_reference_eval_together_inference eval/test_eval.py
```
```
pytest -v -s -m braintrust_scoring_together_inference scoring/test_scoring.py --env OPENAI_API_KEY=$OPENAI_API_KEY
```
<img width="745" alt="image"
src="https://github.com/user-attachments/assets/68f5cdda-f6c8-496d-8b4f-1b3dabeca9c2">
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.
# What does this PR do?
this PR adds a basic inference adapter to NVIDIA NIMs
what it does -
- chat completion api
- tool calls
- streaming
- structured output
- logprobs
- support hosted NIM on integrate.api.nvidia.com
- support downloaded NIM containers
what it does not do -
- completion api
- embedding api
- vision models
- builtin tools
- have certainty that sampling strategies are correct
## Feature/Issue validation/testing/test plan
`pytest -s -v --providers inference=nvidia
llama_stack/providers/tests/inference/ --env NVIDIA_API_KEY=...`
all tests should pass. there are pydantic v1 warnings.
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [x] Did you read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Was this discussed/approved via a Github issue? Please add a link
to it if that's the case.
- [ ] Did you make sure to update the documentation with your changes?
- [x] Did you write any new necessary tests?
Thanks for contributing 🎉!
# What does this PR do?
Safety provider `inline::meta-reference` is now deprecated. However, we
* aren't checking / printing the deprecation message in `llama stack
build`
* make the deprecated (unusable) provider
So I (1) added checking and (2) made `inline::llama-guard` the default
## Test Plan
Before
```
Traceback (most recent call last):
File "/home/dalton/.conda/envs/nov22/bin/llama", line 8, in <module>
sys.exit(main())
File "/home/dalton/all/llama-stack/llama_stack/cli/llama.py", line 46, in main
parser.run(args)
File "/home/dalton/all/llama-stack/llama_stack/cli/llama.py", line 40, in run
args.func(args)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 177, in _run_stack_build_command
self._run_stack_build_command_from_build_config(build_config)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 305, in _run_stack_build_command_from_build_config
self._generate_run_config(build_config, build_dir)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 226, in _generate_run_config
config_type = instantiate_class_type(
File "/home/dalton/all/llama-stack/llama_stack/distribution/utils/dynamic.py", line 12, in instantiate_class_type
module = importlib.import_module(module_name)
File "/home/dalton/.conda/envs/nov22/lib/python3.10/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1050, in _gcd_import
File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
File "<frozen importlib._bootstrap>", line 1004, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'llama_stack.providers.inline.safety.meta_reference'
```
After
```
Traceback (most recent call last):
File "/home/dalton/.conda/envs/nov22/bin/llama", line 8, in <module>
sys.exit(main())
File "/home/dalton/all/llama-stack/llama_stack/cli/llama.py", line 46, in main
parser.run(args)
File "/home/dalton/all/llama-stack/llama_stack/cli/llama.py", line 40, in run
args.func(args)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 177, in _run_stack_build_command
self._run_stack_build_command_from_build_config(build_config)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 309, in _run_stack_build_command_from_build_config
self._generate_run_config(build_config, build_dir)
File "/home/dalton/all/llama-stack/llama_stack/cli/stack/build.py", line 228, in _generate_run_config
raise InvalidProviderError(p.deprecation_error)
llama_stack.distribution.resolver.InvalidProviderError:
Provider `inline::meta-reference` for API `safety` does not work with the latest Llama Stack.
- if you are using Llama Guard v3, please use the `inline::llama-guard` provider instead.
- if you are using Prompt Guard, please use the `inline::prompt-guard` provider instead.
- if you are using Code Scanner, please use the `inline::code-scanner` provider instead.
```
<img width="469" alt="Screenshot 2024-11-22 at 4 10 24 PM"
src="https://github.com/user-attachments/assets/8c2e09fe-379a-4504-b246-7925f80a6ed6">
## Sources
Please link relevant resources if necessary.
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [x] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.
# What does this PR do?
- move folder
## Test Plan
**Unit Test**
```
pytest -v -s -m "huggingface" datasetio/test_datasetio.py
```
**E2E**
```
llama stack run
```
```
llama-stack-client eval run_benchmark meta-reference-mmlu --num-examples 5 --output-dir ./ --eval-task-config ~/eval_task_config.json --visualize
```
<img width="657" alt="image"
src="https://github.com/user-attachments/assets/63d53f9d-6c7e-4667-af8c-9d16c91ae6e3">
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.
# What does this PR do?
- add inline:: prefix for localfs provider
## Test Plan
```
llama stack run
datasetio:
- provider_id: localfs-0
provider_type: inline::localfs
config: {}
```
```
pytest -v -s -m meta_reference_eval_fireworks_inference eval/test_eval.py
pytest -v -s -m localfs datasetio/test_datasetio.py
```
## Sources
Please link relevant resources if necessary.
## Before submitting
- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.
# What does this PR do?
This PR kills the notion of "pure passthrough" remote providers. You
cannot specify a single provider you must specify a whole distribution
(stack) as remote.
This PR also significantly fixes / upgrades testing infrastructure so
you can now test against a remotely hosted stack server by just doing
```bash
pytest -s -v -m remote test_agents.py \
--inference-model=Llama3.1-8B-Instruct --safety-shield=Llama-Guard-3-1B \
--env REMOTE_STACK_URL=http://localhost:5001
```
Also fixed `test_agents_persistence.py` (which was broken) and killed
some deprecated testing functions.
## Test Plan
All the tests.
Splits the meta-reference safety implementation into three distinct providers:
- inline::llama-guard
- inline::prompt-guard
- inline::code-scanner
Note that this PR is a backward incompatible change to the llama stack server. I have added deprecation_error field to ProviderSpec -- the server reads it and immediately barfs. This is used to direct the user with a specific message on what action to perform. An automagical "config upgrade" is a bit too much work to implement right now :/
(Note that we will be gradually prefixing all inline providers with inline:: -- I am only doing this for this set of new providers because otherwise existing configuration files will break even more badly.)
Added support for structured output in the API and added a reference implementation for meta-reference.
A few notes:
* Two formats are specified in the API: Json schema and EBNF based grammar
* Implementation only supports Json for now
We use lm-format-enhancer to provide the implementation right now but may change this especially because BNF grammars aren't supported by that library.
Fireworks has support for structured output and Together has limited supported for it too. Subsequent PRs will add these changes. We would like all our inference providers to provide structured output for llama models since it is an extremely important and highly sought-after need by the developers.
This PR adds support for Qdrant - https://qdrant.tech/ to be used as a vector memory.
I've unit-tested the methods to confirm that they work as intended.
To run Qdrant
```
docker run -p 6333:6333 qdrant/qdrant
```
* docker compose ollama
* comment
* update compose file
* readme for distributions
* readme
* move distribution folders
* move distribution/templates to distributions/
* rename
* kill distribution/templates
* readme
* readme
* build/developer cookbook/new api provider
* developer cookbook
* readme
* readme
* [bugfix] fix case for agent when memory bank registered without specifying provider_id (#264)
* fix case where memory bank is registered without provider_id
* memory test
* agents unit test
* Add an option to not use elastic agents for meta-reference inference (#269)
* Allow overridding checkpoint_dir via config
* Small rename
* Make all methods `async def` again; add completion() for meta-reference (#270)
PR #201 had made several changes while trying to fix issues with getting the stream=False branches of inference and agents API working. As part of this, it made a change which was slightly gratuitous. Namely, making chat_completion() and brethren "def" instead of "async def".
The rationale was that this allowed the user (within llama-stack) of this to use it as:
```
async for chunk in api.chat_completion(params)
```
However, it causes unnecessary confusion for several folks. Given that clients (e.g., llama-stack-apps) anyway use the SDK methods (which are completely isolated) this choice was not ideal. Let's revert back so the call now looks like:
```
async for chunk in await api.chat_completion(params)
```
Bonus: Added a completion() implementation for the meta-reference provider. Technically should have been another PR :)
* Improve an important error message
* update ollama for llama-guard3
* Add vLLM inference provider for OpenAI compatible vLLM server (#178)
This PR adds vLLM inference provider for OpenAI compatible vLLM server.
* Create .readthedocs.yaml
Trying out readthedocs
* Update event_logger.py (#275)
spelling error
* vllm
* build templates
* delete templates
* tmp add back build to avoid merge conflicts
* vllm
* vllm
---------
Co-authored-by: Ashwin Bharambe <ashwin.bharambe@gmail.com>
Co-authored-by: Ashwin Bharambe <ashwin@meta.com>
Co-authored-by: Yuan Tang <terrytangyuan@gmail.com>
Co-authored-by: raghotham <rsm@meta.com>
Co-authored-by: nehal-a2z <nehal@coderabbit.ai>
This PR makes several core changes to the developer experience surrounding Llama Stack.
Background: PR #92 introduced the notion of "routing" to the Llama Stack. It introduces three object types: (1) models, (2) shields and (3) memory banks. Each of these objects can be associated with a distinct provider. So you can get model A to be inferenced locally while model B, C can be inference remotely (e.g.)
However, this had a few drawbacks:
you could not address the provider instances -- i.e., if you configured "meta-reference" with a given model, you could not assign an identifier to this instance which you could re-use later.
the above meant that you could not register a "routing_key" (e.g. model) dynamically and say "please use this existing provider I have already configured" for a new model.
the terms "routing_table" and "routing_key" were exposed directly to the user. in my view, this is way too much overhead for a new user (which almost everyone is.) people come to the stack wanting to do ML and encounter a completely unexpected term.
What this PR does: This PR structures the run config with only a single prominent key:
- providers
Providers are instances of configured provider types. Here's an example which shows two instances of the remote::tgi provider which are serving two different models.
providers:
inference:
- provider_id: foo
provider_type: remote::tgi
config: { ... }
- provider_id: bar
provider_type: remote::tgi
config: { ... }
Secondly, the PR adds dynamic registration of { models | shields | memory_banks } to the API surface. The distribution still acts like a "routing table" (as previously) except that it asks the backing providers for a listing of these objects. For example it asks a TGI or Ollama inference adapter what models it is serving. Only the models that are being actually served can be requested by the user for inference. Otherwise, the Stack server will throw an error.
When dynamically registering these objects, you can use the provider IDs shown above. Info about providers can be obtained using the Api.inspect set of endpoints (/providers, /routes, etc.)
The above examples shows the correspondence between inference providers and models registry items. Things work similarly for the safety <=> shields and memory <=> memory_banks pairs.
Registry: This PR also makes it so that Providers need to implement additional methods for registering and listing objects. For example, each Inference provider is now expected to implement the ModelsProtocolPrivate protocol (naming is not great!) which consists of two methods
register_model
list_models
The goal is to inform the provider that a certain model needs to be supported so the provider can make any relevant backend changes if needed (or throw an error if the model cannot be supported.)
There are many other cleanups included some of which are detailed in a follow-up comment.
This is just like `local` using `meta-reference` for everything except
it uses `vllm` for inference.
Docker works, but So far, `conda` is a bit easier to use with the vllm
provider. The default container base image does not include all the
necessary libraries for all vllm features. More cuda dependencies are
necessary.
I started changing this base image used in this template, but it also
required changes to the Dockerfile, so it was getting too involved to
include in the first PR.
Working so far:
* `python -m llama_stack.apis.inference.client localhost 5000 --model Llama3.2-1B-Instruct --stream True`
* `python -m llama_stack.apis.inference.client localhost 5000 --model Llama3.2-1B-Instruct --stream False`
Example:
```
$ python -m llama_stack.apis.inference.client localhost 5000 --model Llama3.2-1B-Instruct --stream False
User>hello world, write me a 2 sentence poem about the moon
Assistant>
The moon glows bright in the midnight sky
A beacon of light,
```
I have only tested these models:
* `Llama3.1-8B-Instruct` - across 4 GPUs (tensor_parallel_size = 4)
* `Llama3.2-1B-Instruct` - on a single GPU (tensor_parallel_size = 1)
* fixing safety inference and safety adapter for new API spec. Pinned the llama_models version to 0.0.24 as the latest version 0.0.35 has the model descriptor name changed. I was getting the missing package error during runtime as well, hence added the dependency to requirements.txt
* support Llama 3.2 models in Together inference adapter and cleanup Together safety adapter
* fixing model names
* adding vision guard to Together safety
We should use Inference APIs to execute Llama Guard instead of directly needing to use HuggingFace modeling related code. The actual inference consideration is handled by Inference.
