microsoft/onnxruntime

Pull Requests Commits

Add quick start guide and finalize custom operator solution

Copilot committed 1 year ago

533c1518

Complete solution for ONNX Runtime custom operator issue

Copilot committed 1 year ago

d33e0eed

Copilot committed 1 year ago

5dd11f69

[web] fix support for subgroup (#25649)

fs-eire committed 1 year ago

Verified 2ba36825

Add patch for WebGPU on Android to handle fp16 in uniforms (#25349)

vraspar committed 1 year ago

Verified 7b31a10c

Enable 2bit CPU matmul fallback (#25582)

carzh committed 1 year ago

Verified d5d3b287

[QNN EP] Add support for GatherNd Op in QNN EP (#25635)

quic-ashwshan committed 1 year ago

Verified 539d0ed3

Update qMoE spec to support block quantization (#25641)

tianleiwu committed 1 year ago

Verified 59871e3b

[MIGraphX EP] Fix CreateExecutionProviderFactory with correct struct and change vendor_id (#25625)

Zhaeong committed 1 year ago

Verified 14ca6df1

[CANN] Fix CANN build error (#25627)

bachelor-dou committed 1 year ago

Verified 9ecd9f33

Update MoE and qMoE spec (#25619)

tianleiwu committed 1 year ago

Verified 562760a5

[build] use cross-compile to build macOS x86_64 target for WebGPU (#25617)

fs-eire committed 1 year ago

Verified 3b10e447

Add support for QMoE in CPU (#25558)

apsonawane committed 1 year ago

Verified 4004a15b

Move moving weights to memory to the end of Graph::Resolve() (#25626)

yuslepukhin committed 1 year ago

Verified ca8ee367

Add CUDA implementation of GatherBlockQuantized operator (#25575)

xiaomsft committed 1 year ago

Verified 747b4b02

[build] fix macOS x86_64 cross-compile warning (#25615)

fs-eire committed 1 year ago

Verified 14113082

[webgpu] Apply Flash Attention if sliding window exceeds KV cache length (#25594)

daijh committed 1 year ago

Verified 7cc93cf4

Update macOS target version from 13.3 to 13.4 (#25616)

Changming Sun committed 1 year ago

Verified a120b4bf

[QNN-EP] Resolve VTCM buffer sharing bugs (#25622)

quic-calvnguy committed 1 year ago

Verified 3a1bb09b

[QNN EP] Add ONNX ScatterElements support (#24811)

quic-tirupath committed 1 year ago

Verified f755b8a8

[QNN EP] Bug fix: multiple consumer for cast result in name conflicts (#25584)

qti-yuduo committed 1 year ago

Verified 554cda57

Optimize layout for SubgroupMatrixLoad on Intel (#25384)

jchen10 committed 1 year ago

Verified b77dbd43

[QNN EP] Lower Gemm with 2d bias to FC + ElementwiseAdd when targeting HTP. (#25605)

quic-muchhsu committed 1 year ago

Verified e57dc2a2

[build] disable CodeQL for NPM Packaging Pipeline (#25614)

fs-eire committed 1 year ago

Verified 7b2f667e

Cache opSupportLimits to improve the performance and update tracing e… (#25589)

qwu16 committed 1 year ago

Verified a7bc727a

Refactor Java Test Pipeline (#25608)

Changming Sun committed 1 year ago

Verified ac7af24a

[QNN EP] Add Unit tests for LPBQ Fusions (#25592)

quic-tirupath committed 1 year ago

Verified 5c0a7d81

Disable Turing GPU for Nv Trt Rtx Ep (#25611)

ishwar-raut1 committed 1 year ago

Verified 780c0e17

[CUDA] BF16 MoE and qMoE (#25572)

tianleiwu committed 1 year ago

Verified 68b9d9bf

[VitisAI] add new api to VitisAI to save graph as a string (#25602)

yifei410 committed 1 year ago

Verified 866c7e3a

Older