Skip to content

Fused rms#61

Merged
k50112113 merged 10 commits intoshaoclee/ds_fp4_gemmfrom
omuhamma/dsfp4-rms
Dec 26, 2025
Merged

Fused rms#61
k50112113 merged 10 commits intoshaoclee/ds_fp4_gemmfrom
omuhamma/dsfp4-rms

Conversation

@omuhamma
Copy link
Contributor

@omuhamma omuhamma commented Dec 16, 2025

Motivation

Porting over the fused_rms to ATOM

Test Plan

Checking Trace, Accuracy and Performance

Test Result

Trace shows the correct kernels
There is no accuracy drop-off (around ~94% with or without rms enabled)
Performance will be in an excel sheet

Submission Checklist

@omuhamma omuhamma self-assigned this Dec 17, 2025
@k50112113 k50112113 marked this pull request as ready for review December 26, 2025 15:43
@k50112113 k50112113 merged commit 98fdbb4 into shaoclee/ds_fp4_gemm Dec 26, 2025
4 checks passed
@k50112113 k50112113 deleted the omuhamma/dsfp4-rms branch December 26, 2025 15:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants