lm-sys / lm-sys/FastChat

llama-2 70b model using openai_api_server chat completions api or gradio interface times out

Open
#2,457 5 comments 0 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

trying to have conversation using the gradio interface or openai_api_server with llama-2 70b model ( with. 4 gous ), it kind of times out. has anyone had any success with this? at one point i got this error from gradio interface, curl using openai_api_server chat completions takes more than 3-4mins and never completes

Error -
NETWORK ERROR DUE TO HIGH TRAFFIC. PLEASE REGENERATE OR REFRESH THIS PAGE.

(Tensor on device cuda:0 is not on the expected device meta!)

(error_code: 50001)

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 reproducing the timeout through the gradio interface and the openai_api_server chat completions API with the llama-2 70b model and the reported 4-GPU setup. Inspect the reported CUDA device mismatch and determine whether either entry point completes reliably; done means identifying a reproducible cause and confirming successful chat completions without the timeout or device error.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
api, backend, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.