stop_str not working for vllm_worker
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
Hi I am using vllm_worker to serve my model, `<|im_end|>` had added in the conversation templates `stop_str`, resulted as:
```
stop_str=["<|endoftext|>", "<|im_end|>"]
```
But when calling the api with openai api, the inference results still contains the `<|im_end|>` and not stop inferencing.
```
it.<|im_end|>\\\n\nturn8\nI:
```
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the vllm_worker serving path and its OpenAI API handling of the conversation template's stop_str values. Reproduce the request with the supplied stop strings and verify that generation stops before <|im_end|> is included in the response.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- api, backend
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100