onnxruntime
Fix LinearAttention output shape inference for standard GQA (q > kv)
#29892

Merged

Fix LinearAttention output shape inference for standard GQA (q > kv) #29892

mirounga merged 2 commits into microsoft:main from mirounga:rs_buf71

Fix LinearAttention output shape inference for standard GQA (q > kv)

24f9b884

[Webgpu] Fix LinearAttention output shape for standard GQA (q > kv)

a6acfff9

mirounga assigned

mirounga 11 days ago

hariharans29 requested a review from

copilot-pull-request-reviewer 11 days ago

hariharans29 approved these changes on 2026-07-30

hariharans29 commented on 2026-07-30

copilot-pull-request-reviewer commented on 2026-07-30

mirounga merged b725f3d3 into main 9 days ago

Reviewers

hariharans29

Assignees

mirounga

Labels

None yet

Milestone

No milestone