chore: bump flashinfer version to 0.6.7#21422
Merged
Conversation
sglang-bot
requested review from
CatherineSue,
Fridge003,
HaiShaw,
JustinTong0323,
ishandhanani,
ispobock,
merrymercy,
slin1237 and
yctseng0211
as code owners
Contributor
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
github-actions
bot
added
the
dependencies
Collaborator
|
/tag-and-rerun-ci |
Collaborator
|
/rerun-failed-ci again?? |
Collaborator
|
Piecewise cuda graph failuer: #21452 |
Collaborator
|
RMSNorm failure (?) Seem related https://github.com/sgl-project/sglang/actions/runs/23566614089/job/68672358288?pr=21422 |
This commit updates the flashinfer version across all relevant files:
- docker/Dockerfile
- python/pyproject.toml
- python/sglang/srt/entrypoints/engine.py
- python/sglang/srt/utils/common.py
🤖 Generated with GitHub Actions
b8zhong
force-pushed
the
bot/bump-flashinfer-version-0.6.7-d69d
branch
from
a428d43 to
d804ed9
Compare
Contributor
|
#21625 will fix v0.6.7 flaky |
4 tasks
…6.7 compatibility Flashinfer 0.6.7 switched to CuTe-based kernels with stricter dtype validation for rmsnorm. When weight (fp32) and input (bf16/fp16) dtypes mismatch, the new kernels raise ValueError. Cast weight to input dtype before calling flashinfer norm functions. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fridge003
requested review from
BBuf,
FlamingoPg and
yizhang2077
as code owners
…infer 0.6.7 compatibility" This reverts commit 7356c9d.
Fridge003
requested review from
DarkSharpness,
HydraQYH,
celve and
yuan-luo
as code owners
This was referenced Mar 31, 2026
5 tasks
This was referenced Apr 1, 2026
Fridge003
approved these changes
Apr 1, 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
5 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.
Summary
This PR bumps the flashinfer version to
0.6.7across all relevant files.Fix these bugs:
Fix #19081
Fix #18989
Fix #18980
, this version include this commit: flashinfer-ai/flashinfer#2726
Files Updated
🤖 Generated with GitHub Actions