Skip to content

Comments

Clean up fp8 support#4230

Merged
merrymercy merged 3 commits intomainfrom
pr-cleanup-fp8
Mar 10, 2025
Merged

Clean up fp8 support#4230
merrymercy merged 3 commits intomainfrom
pr-cleanup-fp8

Conversation

@merrymercy
Copy link
Contributor

@merrymercy merrymercy commented Mar 9, 2025

  • move duplicated code for hip into a single function
  • test fp8 on AMD
  • use get_bool_env_var for USE_VLLM_CUTLASS_W8A8_FP8_KERNEL

TODO: Fix compressed tensors on mixtral

@merrymercy merrymercy merged commit e8a69e4 into main Mar 10, 2025
23 checks passed
@merrymercy merrymercy deleted the pr-cleanup-fp8 branch March 10, 2025 04:46
aoshen524 pushed a commit to aoshen524/sglang that referenced this pull request Mar 10, 2025
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