microsoft/onnxruntime

Pull Requests Commits

Add optional dependency for quantization utilities (ml_dtypes) (#27435)

chaya2350 committed 8 days ago

Verified 7c56ee85

Fix 3d attention mask broadcasting in MHA (#27464)

titaiwangms committed 8 days ago

Verified 654c335c

Fix GatherCopyData Integer Truncation Leading to Heap Out-of-Bounds Read/Write (#27444)

chilo-ms committed 8 days ago

Verified 355278a3

Update python_requires (#27354)

XXXXRT666 committed 8 days ago

Verified eb23be8d

Fix SkipLayerNorm fusion incorrectly applied when gamma/beta are not 1D (#27459)

Copilot committed 9 days ago

Verified 82eb99c0

optimized qmoe code path for 1 token (#27383)

guschmue committed 9 days ago

Verified 5f087c41

[wasm] add flags for WebGPU build to reduce binary size (#27371)

fs-eire committed 9 days ago

Verified 3db53eb0

WebGPU: Optimize GraphCacheManager by removing pending_buffers (#27453)

qjia7 committed 9 days ago

Verified c6491e0e

[MLAS] ukernel interface logging changes (#27439)

JonathanC-ARM committed 9 days ago

Verified 98cc06e8

Fix VersionNumber usage in React Native Android build with major version validation (#27385)

Copilot committed 9 days ago

Verified f3a62845

Add OrtModel input support for Compile API (#27332)

adrastogi committed 9 days ago

Verified 500ab5f1

fix: OrtValue.from_dlpack rejects zero-size tensors as non-contiguous (#27451)

Copilot committed 9 days ago

Verified 616a4dfc

webgpu: support head_sink in flash attention (#27410)

guschmue committed 10 days ago

Verified bb3866cf

Improve Pre Packing for 2 bit LUT kernels (#27131)

vraspar committed 10 days ago

Verified 2145c8c0

fix: out of bounds access for resize operation (#27419)

lukas-folle-snkeos committed 10 days ago

Verified 79e0676e

Add validity check on MaxUnpool indicies. (#27432)

skottmckay committed 10 days ago

Verified 1f05a4ce

webgpu: Add configurable multi rotary cache concat offset parameter (#27434)

qjia7 committed 10 days ago

Verified 88855c7f

Enable Python 3.14 CI and Upgrade Dependencies (#27401)

tianleiwu committed 10 days ago

Verified 71d2cd01

Fix validation for external data paths for models loaded from bytes (#27430)

adrianlizarraga committed 10 days ago

Verified 9afb0d2b

[CUDA EP] Add resize op version 19 support for CUDA (#27415)

ShirasawaSama committed 10 days ago

Verified 0e9b5af6

[Build] Fix nuget pipelines to pass IsReleaseBuild to validate-package (#27387)

tianleiwu committed 10 days ago

Verified f1ab2e8b

MatmulNBits prepacking scales fix (#27412)

hariharans29 committed 10 days ago

Verified 09828443

Normalize Release/Non Release DML packaging pipelines (#27395)

eserscor committed 10 days ago

Verified 6e7c5f80

Move QNN publishing to release pipeline solely (#27420)

eserscor committed 10 days ago

Verified 920cf11a

Fix build issues in winml_adapter (#27423)

jnagi-intel committed 11 days ago

Verified 649a3559

Fix DllImportResolver (#27397)

tianleiwu committed 11 days ago

Verified 426b0066

TensorScatter-24 (CPU) (#27389)

titaiwangms committed 11 days ago

Verified a8f34cc3

Fix error where bytes is not assigned for dynamic qgemm pack b size (#27421)

JonathanC-ARM committed 11 days ago

Verified b32b205d

Fix refcount bug in map input conversion that caused shutdown segfault (#27413)

tianleiwu committed 11 days ago

Verified f501e1d0

Support nonpad kv seqlen within opset 24 Attention (CPU) (#27384)

titaiwangms committed 11 days ago

Verified a1b634cf

Newer Older