onnxruntime
DQ→MatMulNBits fusion transformer for NvTensorRtRtx ep
#27466
Merged

DQ→MatMulNBits fusion transformer for NvTensorRtRtx ep #27466

anujj
anujj optimizer: fuse WebNN DQ chain back to MatMulNBits for NvTensorRTRTX
7e665906
anujj optimizer(webnn): keep DQ->MatMulNBits fusion on phase-1 safe path
0434a1f4
anujj optimizer(webnn): restore Gemm bias preserving DQ to MatMulNBits fusi…
e38ae4fa
praneshgo Add SDXL Turbo fusion pattern for QDQ+MatMul -> MatMulNBits fusion
e4332496
anujj optimizer(webnn): harden direct DQ->MatMulNBits fusion shape checks
887b6251
anujj file name changes
f98418ff
anskumar01
xadupre
xadupre commented on 2026-02-26
xadupre
xadupre commented on 2026-02-26
xadupre
xadupre commented on 2026-02-27
anujj refactor: split DQMatMulNBitsFusion::ApplyImpl into focused helpers
8559bcd8
fdwr
github-advanced-security
github-advanced-security commented on 2026-03-02
tianleiwu
tianleiwu tianleiwu added release:1.24.3
anujj anujj force pushed from fbce2c02 to 210a7c41 10 days ago
anujj anujj force pushed from 210a7c41 to 143f7e04 10 days ago
anujj address PR review: config-driven gating, SafeInt, minimal-build guard…
40207a90
anujj anujj force pushed from 143f7e04 to 40207a90 10 days ago
anujj
tianleiwu
tianleiwu approved these changes on 2026-03-04
tianleiwu tianleiwu enabled auto-merge (squash) 9 days ago
tianleiwu
azure-pipelines
tianleiwu tianleiwu merged 5c3f5449 into main 9 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone