NVIDIA / NVIDIA/cutlass

[QST] Does CUTLASS 3.5.1 support int4 x float16 GEMMs natively?

Open
#1,928 5 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

What is your question?
By "natively," I mean without relying on third-party implementations. If I understand correctly, FasterTransformer and TVM have already developed their own CUTLASS extensions for constructing INT4/INT8 x FLOAT16 GEMMs. Just wondering if the latest CUTLASS release can already do this now? Thanks!

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

The issue names no source file, test, or entry point. Start by checking CUTLASS 3.5.1's documented and example GEMM support for INT4/INT8 × FLOAT16, then compare it with the third-party extensions mentioned for FasterTransformer and TVM. Done means confirming whether native support exists or documenting the gap.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
performance
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.