Commits microsoft/onnxruntime

Fix CPU fallback type check and add graph-level coverage

Gopalakrishnan Nallasamy committed 7 days ago

c58c3426

Merge branch 'main' into GopalakrishnanN/constexpr-kernel-capability-types

GopalakrishnanN committed 20 days ago

Verified 23353a58

Fix type/shape inferencing double recursion for subgraphs (#29617)

skottmckay committed 20 days ago

Verified da904943

add timeout to every CI job (default is currently 6h when missing) (#29765)

xadupre committed 20 days ago

Verified 7742d3d4

[Build] install java in Windows nuget packaging (#29774)

tianleiwu committed 20 days ago

Verified 204e0f59

[CPU] Fix output saturation in int8 QLinearSoftmax (#29728)

mcollinswisc committed 21 days ago

Verified 8585c64b

[WebGPU] updates webgpu softmax to use online algorithm (#29694)

prathikr committed 21 days ago

Verified 0c0d9aab

Ignore excess data in DFT. (#29680)

skottmckay committed 21 days ago

Verified a7f1699b

Additional validation in LSTM and DynamicQuantizedLSTM (#29595)

skottmckay committed 21 days ago

Verified fb999b4e

Add OrtErrorCode::ORT_DEVICE_RESET. (#29748)

edgchen1 committed 21 days ago

Verified 5911a3a2

webgpu: add MatMulBnb4 contrib op support (#29587)

xhcao committed 21 days ago

Verified 0656e947

[MLAS] Add RISC-V RVV backend for MLAS QNBitGemm (MatMulNBits) (#29537)

velonica0 committed 21 days ago

Verified d3fab37b

Enable fpA_intB GEMM in CUDA builds and add configurable options (#29622)

tianleiwu committed 21 days ago

Verified 308f24c3

Fix structure of EpLibraryPlugin::Load so that we call Unload on error. (#29634)

skottmckay committed 21 days ago

Verified 263bed35

Add Java setup to Windows x64 QNN CI Pipeline (#29754)

mcollinswisc committed 21 days ago

Verified af9b0a4b

[WebGPU] Upgrade Dawn to the latest (#29389)

Jiawei-Shao committed 21 days ago

Verified bf6aa006

[MLAS][KleidiAI] Add MLAS Arm64 half GEMM and convolution support (#28786)

Laan33 committed 21 days ago

Verified 01f0dba6

Add Intel WebGPU subgroup-matrix MatMul implementation (#29592)

jchen10 committed 21 days ago

Verified 86916eeb

Add Windows ARM64 CUDA plugin package and align CUDA metadata/artifact naming (#28896)

tianleiwu committed 21 days ago

Verified 405fbea3

[MLAS] Add AVX2 (+VNNI) 2-bit weight CPU kernels (#29619)

blazingphoenix7 committed 21 days ago

Verified a4e58e8f

[ORT] Extend reshape fusion (#29027)

Honry committed 21 days ago

Verified 737328ba

[CUDA] Fix null allocator passed to plugin EP kernel PrePack (#29658)

tianleiwu committed 21 days ago

Verified 41bd3911

[BUILD] CUDA_QUANT_PREPROCESS off by default and Adjust CI (#29687)

tianleiwu committed 21 days ago

Verified 1c89b863

Fix CPU AvgPool ceil_mode + count_include_pad wrong average (opset 7-18 / MLAS path) (#29629)

titaiwangms committed 22 days ago

Verified bead9137

Validate GridSample input spatial dimensions are non-empty (#29255)

titaiwangms committed 22 days ago

Verified 7bde5d71

Enable Spectre-mitigated MSVC libs for BinSkim builds (#29624)

tianleiwu committed 22 days ago

Verified f05b2186

Fix CPU FlashAttention disabled on Linux/aarch64 by detecting L2 cache via cpuinfo (#29621)

Sammy-Dabbas committed 22 days ago

Verified 3159e0a1

[WebNN EP] Fix scale/zero_point rank for per-axis QDQ on last axis (#29475)

Honry committed 22 days ago

Verified 192ab239

Detach C++ node tests from ONNX on-disk data/node artifacts (onnx#7959) + retiring equivalence proof + corpus-collapse tripwire (#29504)

titaiwangms committed 22 days ago

Verified c4f76f6f

[CUDA] Add cuDNN-free ArgMax/ArgMin/ReduceSum and fix LogSoftmax on plugin EP (#29620)

tianleiwu committed 22 days ago

Verified fea45a31