microsoft/onnxruntime

Pull Requests Commits

[NV TRT RTX EP] Package CUDA Runtime and TRT RTX with NV TensorRT RTX EP python wheel (#26115)

keshavv27 committed 155 days ago

Verified 5568c3af

Disable build flatc for tvOS or visionOS (#26380)

hans00 committed 155 days ago

Verified 0463aa9f

Add LSX support for S8S8 and S8U8 GEMM kernels (#24397)

wszqkzqk committed 155 days ago

Verified 8267066d

[CANN] Fix the ACL_ERROR_REPEAT_INITIALIZE error that occurs when coexisting… (#26193)

bachelor-dou committed 157 days ago

Verified f20df72d

Add CreateQuantPybindModule into orttraining (#26401)

jiafatom committed 157 days ago

Verified 697283dd

[MLAS] Fix rotary avx2 kernel invalid access (#26389)

tianleiwu committed 157 days ago

Verified 1f458385

Fix for micro benchmark build failure with c++20 (#26299)

JonathanC-ARM committed 157 days ago

Verified 6e676641

FP16 inference performance improvement on CPU (#25680)

nikhilfujitsu committed 157 days ago

Verified 42869cac

[webgpu] Use DoTranspose directly in Conv Op (#26074)

wenqinI committed 159 days ago

Verified b3ba5803

[webgpu] Remove 'webgpu' suffix from filenames (#26287)

daijh committed 159 days ago

Verified cd3aa49d

[webgpu] Fix graph capture broken (#26383)

qjia7 committed 160 days ago

Verified 949c0de7

Bump parkerbxyz/suggest-changes from 2 to 3 (#25839)

dependabot[bot] committed 160 days ago

Verified c7dfc25d

Bump vite from 6.3.6 to 6.4.1 in /js/web/test/e2e/exports/testcases/vite-default (#26370)

dependabot[bot] committed 160 days ago

Verified e7ef489d

[webgpu] Expand Squeeze version to 24 (#26359)

qjia7 committed 160 days ago

Verified b6b90a08

[webgpu] Simplify copyKVCache (#26371)

qjia7 committed 160 days ago

Verified 14f540b7

fix npm publishing (#26305)

guschmue committed 160 days ago

Verified 5ef8ce21

[WebGPU] Fix the typo `splitted` with `split` (#26360)

Jiawei-Shao committed 160 days ago

Verified e30f604b

[QNN-EP] Enable Optrace from QNN into QNN EP (#25987)

quic-calvnguy committed 160 days ago

Verified 50f07e27

CuFFTPlanCache: add destructor and Clear() for memory release (#26098)

cocotdf committed 160 days ago

Verified 50317ffb

Fix onnxruntime build on s390x with more recent GCC versions (#26374)

AlekseiNikiforovIBM committed 160 days ago

Verified 882c301b

[WebGPU] allows user to specify high-performance or low-power preference (#26326)

prathikr committed 160 days ago

Verified 00838cc7

[Web] Magic import comment missing for Vite (/* @vite-ignore */) (#25919)

mcost45 committed 160 days ago

Verified f5654ed8

[Build] New docker image with GCC 14 and CUDA 12.8/13.0 (#26349)

tianleiwu committed 161 days ago

Verified b970c118

Bump actions/setup-node from 5 to 6 (#26357)

dependabot[bot] committed 161 days ago

Verified ae71a573

[web] support JSPI build (#26095)

fs-eire committed 161 days ago

Verified 6efb844d

Fix quantization in Whisper model export (#26353)

jiafatom committed 162 days ago

Verified ed7847c1

[webgpu] Don't use num_workgroups when use indirect dispatch (#26334)

qjia7 committed 162 days ago

Verified 8ab27d9f

[build] disable DAWN_FETCH_DEPENDENCIES when using custom dawn src path (#26346)

fs-eire committed 162 days ago

Verified b1fdf9d3

[LARCH64 CPU]Provide inference acceleration optimization for Loongson CPU with 4-bit quantized models (#26280)

movedancer committed 164 days ago

Verified 7cc28b0b

[QNN EP] Reshape Transpose Fusion for 6D tensors (#26338)

quic-ashwshan committed 164 days ago

Verified 7804b5c1

Newer Older