microsoft/onnxruntime

Pull Requests Commits

Add dynamic-shape memory-pattern eligibility test

Gopalakrishnan Nallasamy committed 5 days ago

10f7dc51

[MLAS] Add ARM64 2-bit weight CPU kernels (#29466)

hariharans29 committed 7 days ago

Verified 5f4538ac

Shard Windows ASan test runs (#29503)

edgchen1 committed 8 days ago

Verified abe7ba92

[EP ABI] Add API to select the best compiled model compatibility info from candidate strings (#28387)

chilo-ms committed 8 days ago

Verified af0588b4

Fix negative-axis handling in ExpandDims shape inference (#29448)

tianleiwu committed 8 days ago

Verified d7efadab

Fix GQA out-of-bounds read in past KV buffer (CPU) (#29447)

tianleiwu committed 8 days ago

Verified 3eb4a356

Fix copy-paste typos in DynamicQuantizeLSTM zero-point and scale validation (#29462)

chilo-ms committed 8 days ago

Verified 99ba5db1

WebGPU: Add indirect dispatch for flash attention graph capture to support Gemma4 (#29236)

feich-ms committed 9 days ago

Verified d7592544

[CUDA] Batched small-M GEMV for MatMulNBits (4-bit/8-bit) (#29451)

jambayk committed 10 days ago

Verified 0b5580da

[OVEP] OpenVINO Development Updates (#28954)

ankitm3k committed 10 days ago

Verified d64a76cc

Fix brew install applesimutils failure by trusting wix/brew tap (#29450)

chilo-ms committed 10 days ago

Verified 36c6b7e1

Harden ConvTranspose pad computation with SafeInt and consistency checks (#29446)

yuslepukhin committed 10 days ago

Verified 6f31cd68

Expand model package with authoring tools and schema versioning (#28989)

jambayk committed 10 days ago

Verified 76936d30

[CUDA] Make cuDNN optional at runtime (#29252)

tianleiwu committed 10 days ago

Verified 8e8190b5

[VitisAI] Recognize native-endian in-memory address tag in process_ext_address (#29248)

mingyueliuh committed 10 days ago

Verified 4f2ba9c4

Validate and inline external data in node tensor attributes during session initialization (#29250)

yuslepukhin committed 11 days ago

Verified 1a98f158

Handle non-trivially-copyable types in Loop/Scan output concatenation (#29397)

yuslepukhin committed 11 days ago

Verified 152d01f7

Fix optional-output guard in DecoderAttention/MultiHeadAttention shape inference (#29268)

titaiwangms committed 11 days ago

Verified a83b776e

Reject CUDA BERT EmbedLayerNorm/SkipLayerNorm shapes exceeding 32-bit output indexing (#29264)

titaiwangms committed 11 days ago

Verified 605a354b

[VSINPU]Fix VSINPU ep build error (#29440)

xuke537 committed 11 days ago

Verified 85cb7f11

[CoreML] Address "model_path" must not be empty (#29394)

yuslepukhin committed 11 days ago

Verified ae857d7a

Validate MaxpoolWithMask kernel rank matches input spatial rank (#29253)

titaiwangms committed 11 days ago

Verified 4750af74

[CUDA]: Split-K2 QMoE SwiGLU GEMV kernel (#29167)

tianleiwu committed 11 days ago

Verified e0b2b970

[CUDA] Deprecate SkipLayerNorm strict mode (#29388)

tianleiwu committed 11 days ago

Verified 7e8fc295

Fix data race in CUDA graph tests: add SynchronizeInputs/SynchronizeOutputs (#29398)

Copilot committed 11 days ago

Verified c47718e2

cuda plugin version 0.1.0 (#29390)

tianleiwu committed 11 days ago

Verified ec86d8db

Fix TreeEnsemble target id validation (#29293)

edgchen1 committed 11 days ago

Verified 3c8c4602

Fix arbitrary memory read (#29011)

apsonawane committed 12 days ago

Verified f285b9e8

Fix SimplifiedLayerNorm fusion with node-produced Pow exponent (#29196)

the0cp committed 12 days ago

Verified 5b6f1e67

Fix Windows Cuda 13.3 build for VS 2026 (#29266)

tianleiwu committed 12 days ago

Verified 4bfb60e6

Older