Skip to content

Bump FlashInfer to 0.6.15 and revert regressions - #31502

Merged
Fridge003 merged 5 commits into
mainfrom
mmangkad/flashinfer-0.6.15-revert-regressions
Jul 17, 2026
Merged

Bump FlashInfer to 0.6.15 and revert regressions#31502
Fridge003 merged 5 commits into
mainfrom
mmangkad/flashinfer-0.6.15-revert-regressions

Conversation

@mmangkad

@mmangkad mmangkad commented Jul 16, 2026 •
edited by github-actions Bot
Loading

Copy link
Copy Markdown
Collaborator

CI States

Latest PR Test (Base): 🚫 Run #29549992666
Latest PR Test (Extra): 🚫 Run #29549992592

@gemini-code-assist

gemini-code-assist Bot commented Jul 16, 2026

Copy link
Copy Markdown
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 github-actions Bot added dependencies Pull requests that update a dependency file deepseek blackwell SM100/SM120 labels Jul 16, 2026
@Fridge003

Fridge003 commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

/rerun-failed-ci

@mmangkad
mmangkad marked this pull request as draft July 17, 2026 01:14
@mmangkad

mmangkad commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator Author

Hold, it seems they deleted the release

@mmangkad

mmangkad commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator Author

Back now

@mmangkad
mmangkad marked this pull request as ready for review July 17, 2026 01:31
@gemini-code-assist

gemini-code-assist Bot commented Jul 17, 2026

Copy link
Copy Markdown
Contributor

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

@nvpohanh

nvpohanh commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator 
  File "/actions-runner/_work/sglang/sglang/python/sglang/srt/layers/quantization/modelopt_quant.py", line 2535, in apply
    ensure_cutedsl_wrapper(layer)
  File "/actions-runner/_work/sglang/sglang/python/sglang/srt/layers/moe/moe_runner/flashinfer_cutedsl.py", line 276, in ensure_cutedsl_wrapper
    layer._cutedsl_wrapper = CuteDslMoEWrapper(
                             ^^^^^^^^^^^^^^^^^^
TypeError: CuteDslMoEWrapper.__init__() got an unexpected keyword argument 'activation'

https://github.com/sgl-project/sglang/actions/runs/29547646260/job/87783429763?pr=31502
🤔

@YAMY1234 YAMY1234 mentioned this pull request Jul 17, 2026
Closed
@b8zhong

b8zhong commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

/rerun-test test/registered/models_e2e/test_dsa_glm52_nvfp4_tp_mtp.py

@github-actions

github-actions Bot commented Jul 17, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Results for /rerun-test test/registered/models_e2e/test_dsa_glm52_nvfp4_tp_mtp.py:

🚀 4-gpu-b200 (1 test): ✅ View workflow run 

cd test/ && python3 registered/models_e2e/test_dsa_glm52_nvfp4_tp_mtp.py

@b8zhong
fix: CuteDslMoEWrapper activation_type kwarg after flashinfer 0.6.15 …
3fc9d79 
…bump

flashinfer-ai/flashinfer#3737 replaced CuteDslMoEWrapper's `activation: str`
kwarg with `activation_type: ActivationType`, breaking construction of the
wrapper on the flashinfer_cutedsl MoE path.

Claude-Session: https://claude.ai/code/session_01HrFNbKCKBijK2QESTD4nt9
@b8zhong

b8zhong commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

@nvpohanh Just fixed that

@b8zhong

b8zhong commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

/rerun-test test/registered/ep/test_flashinfer_a2a.py

@github-actions

github-actions Bot commented Jul 17, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Results for /rerun-test test/registered/ep/test_flashinfer_a2a.py:

🚀 4-gpu-gb300 (1 test): ✅ View workflow run 

cd test/ && python3 registered/ep/test_flashinfer_a2a.py

kpham-sgl
kpham-sgl reviewed Jul 17, 2026
View reviewed changes
Comment thread test/registered/models_e2e/test_dsa_glm52_nvfp4_tp_mtp.py Outdated
tp_size = 4
mem_fraction_static = 0.8
bs_1_speed_thres = 250
bs_1_speed_thres = 300

@kpham-sgl kpham-sgl Jul 17, 2026

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Lets change this to 280

b8zhong
b8zhong reviewed Jul 17, 2026
View reviewed changes
Comment thread test/registered/models_e2e/test_dsa_glm52_nvfp4_tp_mtp.py Outdated
@Fridge003
Fridge003 merged commit d67aa05 into main Jul 17, 2026
311 of 353 checks passed
@Fridge003
Fridge003 deleted the mmangkad/flashinfer-0.6.15-revert-regressions branch July 17, 2026 07:50
@b8zhong b8zhong mentioned this pull request Jul 17, 2026
Closed
@Fridge003 Fridge003 mentioned this pull request Jul 17, 2026
Merged
@amd-bot amd-bot mentioned this pull request Jul 20, 2026
Open
@mmangkad mmangkad mentioned this pull request Jul 21, 2026
Merged
Chronostasys pushed a commit to MindLab-Research/sglang that referenced this pull request Aug 24, 2026
@mmangkad @b8zhong
Bump FlashInfer to 0.6.15 and revert regressions (sgl-project#31502)
a56b0d4 
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
jakki-amd pushed a commit to jakki-amd/sglang that referenced this pull request Sep 9, 2026
@mmangkad @b8zhong @jakki-amd
Bump FlashInfer to 0.6.15 and revert regressions (sgl-project#31502)
3eb6bbf 
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
@b8zhong b8zhong mentioned this pull request Sep 9, 2026
Open
2 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@b8zhong b8zhong b8zhong left review comments
@kpham-sgl kpham-sgl kpham-sgl left review comments
@fzyzcjy fzyzcjy Awaiting requested review from fzyzcjy fzyzcjy is a code owner
@ispobock ispobock Awaiting requested review from ispobock ispobock is a code owner
@ch-wan ch-wan Awaiting requested review from ch-wan ch-wan is a code owner
@merrymercy merrymercy Awaiting requested review from merrymercy merrymercy is a code owner
@Fridge003 Fridge003 Awaiting requested review from Fridge003 Fridge003 is a code owner
@CatherineSue CatherineSue Awaiting requested review from CatherineSue CatherineSue is a code owner
@slin1237 slin1237 Awaiting requested review from slin1237 slin1237 is a code owner
@JustinTong0323 JustinTong0323 Awaiting requested review from JustinTong0323 JustinTong0323 is a code owner
@HaiShaw HaiShaw Awaiting requested review from HaiShaw HaiShaw is a code owner
@ishandhanani ishandhanani Awaiting requested review from ishandhanani ishandhanani is a code owner
@yctseng0211 yctseng0211 Awaiting requested review from yctseng0211 yctseng0211 is a code owner
@sogalin sogalin Awaiting requested review from sogalin sogalin is a code owner
@Ying1123 Ying1123 Awaiting requested review from Ying1123 Ying1123 is a code owner
@BBuf BBuf Awaiting requested review from BBuf BBuf is a code owner
@Edwardf0t1 Edwardf0t1 Awaiting requested review from Edwardf0t1 Edwardf0t1 is a code owner

Assignees

No one assigned

Labels

blackwell SM100/SM120 bypass-fastfail deepseek dependencies Pull requests that update a dependency file high priority run-ci run-ci-extra

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@mmangkad @Fridge003 @nvpohanh @b8zhong @kpham-sgl