Repository navigation
Conversation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
aryatsc
requested review from
Alisehen,
AniZpZ,
BBuf,
DarkSharpness,
Edwardf0t1,
FlamingoPg,
Fridge003,
HaiShaw,
HydraQYH,
OrangeRedeng,
Ying1123,
b8zhong,
celve,
ch-wan,
ispobock,
merrymercy and
yuan-luo
as code owners
October 26, 2026 20:01
4 of 5 tasks
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> # Conflicts: # python/sglang/srt/environ.py
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> # Conflicts: # python/sglang/kernels/ops/moe/moe_sorting_small.py
This comment has been minimized.
This comment has been minimized.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> # Conflicts: # python/sglang/kernels/ops/moe/__init__.py
This comment has been minimized.
This comment has been minimized.
…erCore.run Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
K3's dense layers run bf16, or dynamic-FP8 with per-step activation scaling.
Modifications
SGLANG_K3_TARGET_DENSE_FP8(defaultoff) loads static-input-scale FP8 dense weights.SGLANG_DRAFT_FP8_ACTIVATION_SCHEMEdoes the same for the DFlash draft.Accuracy Tests
test_fp8_online_static_input.pypasses.Full K3 stack on MI355X TP8 (DFlash, prod config) vs B300, McNemar:
Speed Tests and Profiling
Not measured on its own. It is included in the full-stack K3 MI355X numbers in the other K3 PRs.
Dependencies
kimi_k3.pyedits). Only the last commit is new here.Checklist
CI States
Latest PR Test (Base): ❌ Run #42125645010
Latest PR Test (Extra): ❌ Run #42125644899
Latest PR Test (AMD ROCm 10): ❌ Run #42125644998
Tip
Base CI only covers the core cases, and its workload may not exercise your change. Use
/rerun-testto run the tests your change touches, or copy this prompt for an agent: