NVIDIA / NVIDIA/cutlass

[QST] How to implement a fused mixed precision matrix multiplication such as w4a4 + w16a16?

Open
#2,058 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

? - Needs Triage inactive-30d inactive-90d question
Dominant language
C++
Stars
10.5k
Forks
2.1k
Avg merge
3d 11h
Merged PRs (30d)
7

Description

Dear Team,

I wish to implement a fused mixed precision matrix multiplication such as w4a4 + w16a16 where the w16a16 part is small. An example of this kernel used is for accelerating an LLM with LoRA applied.

I can find some examples in "torchao" that implement matrix multiplication of w4a4/w4a8 and integrate matrix multiplication and dequantization via epilogue, but I don't know how to further integrate matrix multiplication of w16a16 on top of it, is there any examples I can refer to?

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 the torchao examples for w4a4 and w4a8, especially how matrix multiplication and dequantization are combined through an epilogue. Then investigate how a small w16a16 operation for the LoRA use case could be integrated. The issue names no CUTLASS files, tests, entry points, or concrete completion criterion.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, python
Domain
machine-learning, performance
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.