Skip to content

Add FP8 PyTorch inference#3

Merged
WilliamZhang20 merged 11 commits intomainfrom
torch
Dec 22, 2025
Merged

Add FP8 PyTorch inference#3
WilliamZhang20 merged 11 commits intomainfrom
torch

Conversation

@WilliamZhang20
Copy link
Owner

@WilliamZhang20 WilliamZhang20 commented Dec 20, 2025

Attempted stationary tiling of matmuls, but that didn't seem to work.
So far, manufacturability is not good.

Otherwise, Torch inference is solid. Shuffling the test set is also implemented now.
Overall simplified scaling compared to the previous INT8 inference in the torch compiler.

@WilliamZhang20 WilliamZhang20 merged commit 3716f15 into main Dec 22, 2025
6 checks passed
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.

1 participant