Skip to content

[Klaud Cold] minimaxm31flash-fp4-mi355x-sglang-agentic-dspark: disable sparse index top-k sharing (SGLANG_MINIMAX_M3_INDEX_TOPK_FREQ=1) #4469

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Oseltamivir merged 3 commits into from
Oct 25, 2026
Merged

[Klaud Cold] minimaxm31flash-fp4-mi355x-sglang-agentic-dspark: disable sparse index top-k sharing (SGLANG_MINIMAX_M3_INDEX_TOPK_FREQ=1) #4469

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
345c7d3
minimaxm31flash-fp4-mi355x-sglang-agentic-dspark: disable sparse inde…
Oseltamivir Oct 24, 2026
994b3a4
perf-changelog: link #4469
Oseltamivir Oct 24, 2026
e71a406
Merge branch 'main' into exp/minimaxm31flash-mi355x-sgl-topk-freq1
Oseltamivir Oct 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions ...nchmarks/single_node/srt-slurm-recipes/minimaxm31flash/sglang/mi355x-fp4-mtp/agentic.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,9 @@ base:
PYTHONUNBUFFERED: '1'
SGLANG_TIMEOUT_KEEP_ALIVE: '1800'
SGLANG_USE_AITER: '1'
# Recompute the sparse index top-k in every layer (SGLang's ROCm default 2 shares
# one selection across each pair of sparse layers).
SGLANG_MINIMAX_M3_INDEX_TOPK_FREQ: '1'
benchmark:
env:
# Warmup can leave bytes unacknowledged past AIPerf's 30 s default.
Expand Down
6 changes: 6 additions & 0 deletions inferencex-e2e/perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11101,6 +11101,12 @@
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4466
scenario-type: [agentic-coding]

- config-keys:
- minimaxm31flash-fp4-mi355x-sglang-agentic-dspark
description:
- "Set SGLANG_MINIMAX_M3_INDEX_TOPK_FREQ=1 on MI355X SGLang MiniMax-M3.1-Flash AgentX: every sparse layer computes its own index top-k instead of SGLang's ROCm default of sharing one across each pair of layers (sgl-project/sglang#36527)."
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/4469

- config-keys:
- glm5.2-fp8-mi325x-sglang-agentic-mtp
description:
Expand Down
Loading