microsoft/onnxruntime

Pull Requests Commits

Set Java logging level to FINE for debugging JAR testing

adrianlizarraga committed 66 days ago

e325f988

Initial change to enable WebGPU EP to suport plugin API (#25705)

fs-eire committed 66 days ago

Verified 94e7a1c6

Fix DoS vulnerability in FuseReluClip: handle empty tensor (#26878)

tianleiwu committed 67 days ago

Verified 7ad0de5b

Add blackwell check in TRTRTX EP unit tests for FP4/FP8 Custom ops (#26832)

vishalpandya1990 committed 67 days ago

Verified 1a112a28

[webgpu] Fix errors found by matmul2bits tests (#26880)

qjia7 committed 67 days ago

Verified aab56616

[webgpu] make PIX frame generator per session (#26873)

fs-eire committed 67 days ago

Verified 8158d4ed

Do not run auditwheel on QNN whls (#26837)

eserscor committed 68 days ago

Verified 7b9de5b4

[webgpu] refactor initialization of WebGPU Context params (#26855)

fs-eire committed 68 days ago

Verified 2f0eb77b

WebGPU: Use dedicated prepack allocator for kernel prepacking (#26857)

jchen10 committed 70 days ago

Verified 38355ba0

[EP ABI] Add weight pre-packing support to kernel-based plugin EPs (#26754)

adrianlizarraga committed 70 days ago

Verified 2d1ed5b7

Proposed changes to model compilation telemetry (#26804)

adrastogi committed 74 days ago

Verified 573d395d

Add Visual Studio 2026 CMake generator support (#26802)

Copilot committed 74 days ago

Verified ed89ce7c

fix crash in webgpu init when OrtEnv is re-created (#26836)

guschmue committed 75 days ago

Verified 7b5a93ef

[WebGPU] Fix Conv prepacking mismatch with Im2ColMatMul path detection (#26833)

jchen10 committed 75 days ago

Verified 49002015

[QNN-EP] Update gather op input tensor cast logic. (#26835)

quic-muchhsu committed 78 days ago

Verified 925b90b8

Address DML crashes in WebNN QDQ subgraph tests (#26822)

adrastogi committed 78 days ago

Verified 9ce61be1

[QNN EP] Fix Clip op with min or max from QDQ (#26601)

qti-yuduo committed 78 days ago

Verified 9f959081

Bump actions/cache from 4 to 5 (#26795)

dependabot[bot] committed 79 days ago

Verified 07724d5c

[QNN-EP] Restrict zero pads validation check to NPU backend (#26534)

qti-mattsinc committed 79 days ago

Verified 6064a2c1

Bump actions/upload-artifact from 5 to 6 (#26794)

dependabot[bot] committed 79 days ago

Verified 26643ea9

Bump actions/download-artifact from 6 to 7 (#26793)

dependabot[bot] committed 79 days ago

Verified b8ea5971

[ROCM] Add back --rocm_version and --rocm_home (#26819)

tianleiwu committed 79 days ago

Verified 21328f6f

[QNN EP] Upgrade QNN to 2.41.0 (#26812)

qti-jkilpatrick committed 80 days ago

Verified 31c28ae6

WebGPU: Transpose Conv kernels in Prepack (#26675)

jchen10 committed 80 days ago

Verified c2ca4229

Normalize protobuf integration from 4.52 to 6.33 (#26817)

eserscor committed 80 days ago

Verified 19ee5716

[Quantization] Fix static quantize runner usage. (#26624)

minfhong-qti committed 80 days ago

Verified 8e52f390

Add reducemin, reducemax nodes before the 1st consumer of tensor (#26768)

rM-planet committed 80 days ago

Verified da68aa89

[QNN_EP] ConvTranspose not calculating "pad" if "output_shape" is given. (#26665)

quic-ankus committed 80 days ago

Verified 5a5c8913

Fix ExpandDims axis handling bug. (#26805)

edgchen1 committed 80 days ago

Verified 9eb97e73

fix npm packaging pipeline (#26810)

guschmue committed 80 days ago

Verified 64227b58

Older