lm-sys / lm-sys/FastChat

why use --load-8bit is more slow than not add

Open
#2,692 0 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

when run
python -m fastchat.serve.model_worker --model-path chatglm2-6b --num-gpus 2 --load-8bit

generation tokens is more slow than

python -m fastchat.serve.model_worker --model-path chatglm2-6b --num-gpus 2

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 two python -m fastchat.serve.model_worker commands with chatglm2-6b, --num-gpus 2, and the differing --load-8bit option. Inspect the model_worker entry point and compare generation speed and the 8-bit loading path; done means identifying why the option is slower and recording a verified fix or explanation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai
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.