Support guided decoding with vllm and remote::vllm #391
Assignees
Labels
good first issue
Good for newcomers
Comments
|
Hi @ashwinb, @yanxi0830 - I came across this issue and I'm interested in contributing. Do you mind if I take a stab? I've reviewed the contributing guidelines and the existing codebase, and I believe I have a good understanding of what needs to be done. However, if there are any specific details or guidelines you think I should be aware of, please lmk. |
|
@aidando73 Please go ahead. I remember seeing a PR about this (but it was not quite on the correct track) so I think you are good to do. Re: guidelines, consider the Finally, when everything is ready, you'd need to run the tests inside |
|
Thanks @ashwinb 🙏, will get started |
|
@aidando73 Thanks! Please feel free to tag me for review once it's ready. |
3 tasks
aidando73
added a commit
to aidando73/llama-stack
that referenced
this issue
Nov 26, 2024
|
Ok @ashwinb @terrytangyuan PR is ready for review: #528 👈 . Lmk if I've missed anything. |
ashwinb
pushed a commit
that referenced
this issue
Dec 8, 2024
# What does this PR do? Addresses issue (#391) - Adds json structured output for vLLM - Enables structured output tests for vLLM > Give me a recipe for Spaghetti Bolognaise: ```json { "recipe_name": "Spaghetti Bolognaise", "preamble": "Ah, spaghetti bolognaise - the quintessential Italian dish that fills my kitchen with the aromas of childhood nostalgia. As a child, I would watch my nonna cook up a big pot of spaghetti bolognaise every Sunday, filling our small Italian household with the savory scent of simmering meat and tomatoes. The way the sauce would thicken and the spaghetti would al dente - it was love at first bite. And now, as a chef, I want to share that same love with you, so you can recreate these warm, comforting memories at home.", "ingredients": [ "500g minced beef", "1 medium onion, finely chopped", "2 cloves garlic, minced", "1 carrot, finely chopped", " celery, finely chopped", "1 (28 oz) can whole peeled tomatoes", "1 tbsp tomato paste", "1 tsp dried basil", "1 tsp dried oregano", "1 tsp salt", "1/2 tsp black pepper", "1/2 tsp sugar", "1 lb spaghetti", "Grated Parmesan cheese, for serving", "Extra virgin olive oil, for serving" ], "steps": [ "Heat a large pot over medium heat and add a generous drizzle of extra virgin olive oil.", "Add the chopped onion, garlic, carrot, and celery and cook until the vegetables are soft and translucent, about 5-7 minutes.", "Add the minced beef and cook until browned, breaking it up with a spoon as it cooks.", "Add the tomato paste and cook for 1-2 minutes, stirring constantly.", "Add the canned tomatoes, dried basil, dried oregano, salt, black pepper, and sugar. Stir well to combine.", "Bring the sauce to a simmer and let it cook for 20-30 minutes, stirring occasionally, until the sauce has thickened and the flavors have melded together.", "While the sauce cooks, bring a large pot of salted water to a boil and cook the spaghetti according to the package instructions until al dente. Reserve 1 cup of pasta water before draining the spaghetti.", "Add the reserved pasta water to the sauce and stir to combine.", "Combine the cooked spaghetti and sauce, tossing to coat the pasta evenly.", "Serve hot, topped with grated Parmesan cheese and a drizzle of extra virgin olive oil.", "Enjoy!" ] } ``` Generated with Llama-3.2-3B-Instruct model - pretty good for a 3B parameter model 👍 ## Test Plan `pytest -v -s llama_stack/providers/tests/inference/test_text_inference.py -k llama_3b-vllm_remote` With the following setup: ```bash # Environment export INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct export INFERENCE_PORT=8000 export VLLM_URL=http://localhost:8000/v1 # vLLM server sudo docker run --gpus all \ -v $STORAGE_DIR/.cache/huggingface:/root/.cache/huggingface \ --env "HUGGING_FACE_HUB_TOKEN=$(cat ~/.cache/huggingface/token)" \ -p 8000:$INFERENCE_PORT \ --ipc=host \ --net=host \ vllm/vllm-openai:v0.6.3.post1 \ --model $INFERENCE_MODEL # llama-stack server llama stack build --template remote-vllm --image-type conda && llama stack run distributions/remote-vllm/run.yaml \ --port 5001 \ --env INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct ``` Results: ``` llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_model_list[llama_3b-vllm_remote] PASSED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_completion[llama_3b-vllm_remote] SKIPPED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_completions_structured_output[llama_3b-vllm_remote] SKIPPED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_chat_completion_non_streaming[llama_3b-vllm_remote] PASSED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_structured_output[llama_3b-vllm_remote] PASSED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_chat_completion_streaming[llama_3b-vllm_remote] PASSED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_chat_completion_with_tool_calling[llama_3b-vllm_remote] PASSED llama_stack/providers/tests/inference/test_text_inference.py::TestInference::test_chat_completion_with_tool_calling_streaming[llama_3b-vllm_remote] PASSED ================================ 6 passed, 2 skipped, 120 deselected, 2 warnings in 13.26s ================================ ``` ## Sources - vllm-project/vllm#8300 - By default, vLLM uses https://github.com/dottxt-ai/outlines for structured outputs [[1](https://github.com/vllm-project/vllm/blob/32e7db25365415841ebc7c4215851743fbb1bad1/vllm/engine/arg_utils.py#L279-L280)] ## Before submitting [N/A] 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. - [x] Read the [contributor guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md), Pull Request section? [N/A?] Updated relevant documentation. Couldn't find any relevant documentation. Lmk if I've missed anything. - [x] Wrote necessary unit or integration tests.
|
We can close this? |
|
Yes I think so |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
🚀 The feature, motivation and pitch
(fireworks, together, meta-reference) support guided decoding (specifying a json-schema for example, as a "grammar" for decoding) with inference. vLLM supports this functionality -- enable that in the API
Alternatives
No alternatives, this is a core feature that must be supported by all providers (as far as possible).
Additional context
No response
The text was updated successfully, but these errors were encountered: