Optimize MiniMax-H3 - #1334
Merged
Merged
Conversation
Support per-instance dynamic sparse attention settings for the H3 transformer and token refiner. Add an RTX 5090 8-GPU Ulysses preset with Triton pre/post, tensor fusion, head parallel, and FP8 communication.
STwangyingrui
force-pushed
the
yr/opt-h3
branch
from
August 6, 2026 04:42
74d206d to
9d1d351
Compare
helloyongyang
approved these changes
Aug 6, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds configurable dynamic sparse attention for MiniMax-H3 transformer and refiner layers.
It also provides an optimized 8-GPU RTX 5090 preset using Ulysses, Triton pre/post processing, head parallelism, FP8 communication, and tensor fusion.
Note: Sparse attention requires auxiliary training to maintain generation quality.