Add 4x/8x H100/H200 support and FP4 for B200 - #15
Open
deepmancore wants to merge 1 commit into
Open
Conversation
- Tensor-parallel WanModel for sp_size > 2 (existing SP path is hardcoded to <= 2). Enables 4x/8x H100/H200. - NVFP4 W4A4 GEMM for Blackwell (fp4_gemm.py), auto-enabled on compute_cap >= 10.0, FP8 fallback elsewhere. Enables B200. - Bugfixes in model_memory_sp.py: kv_cache dtype crash with fp8_kv_cache=True, and SAGE_FP8_SM90 hardcode breaking non-Hopper GPUs. Measured on H100 SXM (32 frames per chunk @ 24 fps target = 1.333s of video): - 4x H100 SXM: 1.1 s/chunk -> ~29 fps (1.21x realtime) - 8x H100 SXM: 0.9 s/chunk -> ~35 fps (1.48x realtime)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Measured on H100 SXM (32 frames per chunk @ 24 fps target = 1.333s of video):