deepspeedai / deepspeedai/DeepSpeed

[REQUEST] DeepSpeed-FastGen AWQ support

Open
#4,622 3 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
43.1k
Forks
5k
Avg merge
4d 15h
Merged PRs (30d)
112

Description

Is your feature request related to a problem? Please describe.
Without 4bit quantization the batch size is limited

Describe the solution you'd like
Add AWQ support, just like TGI

Describe alternatives you've considered
other 4bit quantization, but AWQ is so far best

Additional context
Add any other context or screenshots about the feature request here.

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.

Research direction

The request names DeepSpeed-FastGen and TGI but does not identify files, tests, or an entry point. Start by locating the DeepSpeed-FastGen inference implementation and reviewing how TGI handles AWQ. Done means AWQ 4-bit quantization is supported and the resulting batch-size limitation is addressed.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.