lm-sys / lm-sys/FastChat

[Feat Request] Adding 72b/110b llava to multimodal chatbot areana

Open
#3,406 0 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

I saw chatbot areana had the multimodal version.

Could we add 72B/110B LLaVA-Next model to it? Or we already had it?
image

We really wish to see the performance on real arena!

HF Collections: https://huggingface.co/collections/lmms-lab/llava-next-6623288e2d61edba3ddbf5ff

I integrated them in SGLang so I think it could be easy to launch it as a backend service.
https://github.com/sgl-project/sglang/blob/main/examples/usage/llava/http_qwen_llava_test.py

Contributor guide

No contributing guide indexed for this repository

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.

Research direction

Start by reviewing FastChat's multimodal chatbot arena entry points and the referenced SGLang example, examples/usage/llava/http_qwen_llava_test.py. Check whether the 72B and 110B LLaVA-Next models are already supported, then define the integration and serving requirements; done means they are available in the real arena for performance evaluation.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, python
Domain
ai
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.