NVIDIA / NVIDIA/Megatron-LM

Support for MXFP8 on 12.0+ architectures

Open
#4,175 2 comments 0 reactions 1 assignee Claimed by @sbhavani View on GitHub
community-request enhancement module: transformer engine waiting-on-maintainers
Dominant language
Python
Stars
17.9k
Forks
4.5k
Avg merge
4d 6h
Merged PRs (30d)
271

Description

**Is your feature request related to a problem? Please describe.**
Hi [@mcore-oncall](https://github.com/orgs/NVIDIA/teams/mcore-oncall) , an error occurs when I use FMXP8 on RTX PRO 6000:
```
AssertionError: MXFP8 (for all gemm layouts) is not supported on 12.0+ architectures yet.
```
Do you have a plan to support MXFP8 on 12.0+ architectures? If the answer is yes, could you please share the planned timeline for it?

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.