onnxruntime
Enable LayerNormFusion for the WebGPU EP to fix fp16 inference correctness
#32294
Merged

Enable LayerNormFusion for the WebGPU EP to fix fp16 inference correctness #32294

Lapis0x0
Lapis0x0 Lapis0x0 requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 23 days ago
azure-pipelines
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-08-27
Lapis0x0
RobertoReale
justinchuby justinchuby requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 22 days ago
github-advanced-security
github-advanced-security commented on 2026-08-27
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2026-08-27
Lapis0x0
RobertoReale
RobertoReale
qjia7
qjia7 approved these changes on 2026-09-03
qjia7 qjia7 requested a review from hariharans29 hariharans29 16 days ago
Lapis0x0
Lapis0x0 Enable LayerNormFusion for the WebGPU EP
cf53a32f
Lapis0x0 Add fp16 decomposed LayerNorm test model
b7d88242
Lapis0x0 Add generator for the fp16 decomposed LayerNorm test model
8af9d142
Lapis0x0 Add LayerNormFusionWebGpu regression test
e42fdbef
Lapis0x0 Preserve fp16 epsilon in LayerNorm fusion
e546e710
Lapis0x0 Lapis0x0 force pushed from e11b2954 to e546e710 16 days ago
Lapis0x0
hariharans29
hariharans29 approved these changes on 2026-09-03
hariharans29 hariharans29 enabled auto-merge (squash) 15 days ago
hariharans29 hariharans29 merged e75f6c89 into main 15 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone