onnxruntime
webgpu: Increase MatMulNBits K-parallelism with tile_size_k_vec=32
#27834
Merged

webgpu: Increase MatMulNBits K-parallelism with tile_size_k_vec=32 #27834

guschmue merged 3 commits into main from webgpu-matmulnbits-step1-correct
qjia7
qjia7 qjia7 marked this pull request as ready for review 15 days ago
qjia7 qjia7 requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 15 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-03-25
qjia7 webgpu: Increase MatMulNBits K-parallelism with tile_size_k_vec=32
304383d3
qjia7 qjia7 force pushed from 183b6e3c to 304383d3 15 days ago
guschmue
guschmue dismissed these changes on 2026-03-25
guschmue guschmue added ep:WebGPU
qjia7 qjia7 dismissed their stale review via 952a4de3 13 days ago
guschmue
azure-pipelines
qjia7 Merge branch 'main' into webgpu-matmulnbits-step1-correct
aada144f
qjia7 qjia7 force pushed from 952a4de3 to aada144f 11 days ago
qjia7 webgpu: Add tile_size_k_vec to MatMulNBits CacheHint
8cd8454c
guschmue
guschmue approved these changes on 2026-03-30
guschmue
guschmue approved these changes on 2026-03-30
guschmue guschmue merged 358628a8 into main 10 days ago
guschmue guschmue deleted the webgpu-matmulnbits-step1-correct branch 10 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone