Comfy-Org/wheels
1
0
Fork 0
mirror of https://github.com/Comfy-Org/wheels.git synced 2026-11-05 00:17:42 +00:00
Code Issues 7 Projects Packages Wiki Activity

[GH-ISSUE #22] SageAttention cu130 / Torch 2.10 wheel missing SM120 architecture #1

New issue
Open
opened 2026-10-26 22:02:08 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-26 22:02:08 +00:00
Owner
Copy link

Originally created by @ironfisc-source on GitHub (Aug 25, 2026).
Original GitHub issue: https://github.com/Comfy-Org/wheels/issues/22

Environment: Windows 11, RTX 5070 Ti (compute capability 12.0), Python 3.13.11, PyTorch 2.10.0+cu130.

The current sageattention-2.2.0+cu130torch2.10-cp313-cp313-win_amd64.whl fails on Blackwell with cudaErrorNoKernelImageForDevice for longer attention workloads.

packages/sageattention.yml has global arch_list: "8.0 9.0". The cu130/Torch 2.9 and Torch 2.11 combinations override this with "8.0 9.0 10.0 12.0", but the cu130/Torch 2.10 combination does not.

Inspecting the installed _fused.cp313-win_amd64.pyd and _qattn_sm89.cp313-win_amd64.pyd shows only sm_80 and sm_90a markers, with no sm_120/sm_120a.

Minimal reproduction with GQA 32:8, head_dim=128, L=4096 produces CUDA error: no kernel image is available for execution on the device, while L=128 works.

Adding arch_list: "8.0 9.0 10.0 12.0" to the cu130/Torch 2.10 matrix entry and rebuilding the wheel appears to be the expected fix.

Originally created by @ironfisc-source on GitHub (Aug 25, 2026). Original GitHub issue: https://github.com/Comfy-Org/wheels/issues/22 Environment: Windows 11, RTX 5070 Ti (compute capability 12.0), Python 3.13.11, PyTorch 2.10.0+cu130. The current sageattention-2.2.0+cu130torch2.10-cp313-cp313-win_amd64.whl fails on Blackwell with cudaErrorNoKernelImageForDevice for longer attention workloads. packages/sageattention.yml has global arch_list: "8.0 9.0". The cu130/Torch 2.9 and Torch 2.11 combinations override this with "8.0 9.0 10.0 12.0", but the cu130/Torch 2.10 combination does not. Inspecting the installed _fused.cp313-win_amd64.pyd and _qattn_sm89.cp313-win_amd64.pyd shows only sm_80 and sm_90a markers, with no sm_120/sm_120a. Minimal reproduction with GQA 32:8, head_dim=128, L=4096 produces CUDA error: no kernel image is available for execution on the device, while L=128 works. Adding arch_list: "8.0 9.0 10.0 12.0" to the cu130/Torch 2.10 matrix entry and rebuilding the wheel appears to be the expected fix.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
dependabot/github_actions/actions/setup-python-7.0.0
dependabot/github_actions/actions/checkout-7.0.1
dependabot/github_actions/actions/attest-build-provenance-4.1.1
dependabot/github_actions/peaceiris/actions-gh-pages-4.1.0
gh-pages
develop
torch_generic_nms-latest
sageattn3-latest
sageattention-latest
cc_torch-latest
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
Comfy-Org/wheels#1
No description provided.