onnxruntime
[CUDA] Enable native SM90, block_size=32, and fused bias for fpA_intB MatMulNBits
#29585

Merged

[CUDA] Enable native SM90, block_size=32, and fused bias for fpA_intB MatMulNBits #29585

tianleiwu merged 1 commit into main from tlwu/20260707/fpa_intb_sm90_bs32_bias

tianleiwu requested a review from

copilot-pull-request-reviewer 3 days ago

copilot-pull-request-reviewer commented on 2026-07-07

[CUDA] Enable native SM90, block_size=32, and fused bias for fpA_intB…

7023483d

tianleiwu force pushed from e3dda06a to 7023483d 3 days ago

jiafatom approved these changes on 2026-07-07

tianleiwu merged e0ad0718 into main 3 days ago

tianleiwu deleted the tlwu/20260707/fpa_intb_sm90_bs32_bias branch 3 days ago

Reviewers

jiafatom

Assignees

No one assigned

Labels

None yet

Milestone

No milestone