Stale To-Do in vLLM/app.py for supporting prompt + generation token IDs

Open Beginner friendly
#1,108 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
2/5
Estimated time
1-3 hours
Newbie friendliness
65/100
Issue type
Feature
Clarity
Clearly specified
Activity status
Quiet
Tech stack
python
Domain
ai

Research direction

Start in responses_api_models/vllm_model/app.py at line 251 and compare the current vLLM and NeMo RL versions with the commented token-ID options. Check how return_token_id_information is handled and run the relevant model tests if available. Done means the stale TODO is addressed and the requested prompt and generation token ID support is wired through.

Written by the indexing model from the issue text.

Description

models

In responses_api_models/vllm_model/app.py, on line 251, there is a block of code that has a TODO regarding adding support for getting generation and prompt token IDs using the flag return_tokens_ids=True. The TODO mentions the upgrade should be done once NeMo RL upgrades to vLLM 0.10.2. NeMo RL is already on 0.17.1.

if self.config.return_token_id_information:
            body_dict |= dict(
                logprobs=True,
                # Typically passed via OpenAI client extra_body.
                return_tokens_as_token_ids=True,
                # TODO add this when NeMo RL upgrades to vLLM 0.10.2 support for prompt token ids
                # For prompt and generation token IDs
                # return_token_ids=True,
                # For prompt token IDs
                # prompt_logprobs=0,
            )
Dominant language
Python
Stars
1.2k
Forks
349
Avg merge
2d 40m
Merged PRs (30d)
320

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA-NeMo/Gym

All issues in NVIDIA-NeMo/Gym

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.