Commits microsoft/onnxruntime

[wasm] adding back uin4/int4 types in webgpu build

fs-eire committed 17 days ago

adb06b6b

Add/Update telemetry events (#27356)

dabhattimsft committed 17 days ago

Verified bdd672aa

[EP Plugin] use ORT_EP_API_ADAPTER instead of BUILD_WEBGPU_EP_STATIC_LIB (#27540)

fs-eire committed 17 days ago

Verified 01dc9995

Merge ConstantOfShape and Cast in clean_graph optimization (#27482)

Rishi-Dave committed 17 days ago

Verified 9c6b7a48

Add OrtEnv.DisableDllImportResolver to prevent fatal error on resolver conflict (#27535)

tianleiwu committed 17 days ago

Verified df8f4a7d

[WebNN] Improve MultiHeadAttention op implementation (#27494)

Honry committed 17 days ago

Verified fe4b6550

[EP API] support OpKernelInfo::GetConfigOptions for adapter (#27541)

fs-eire committed 18 days ago

Verified 763abb9d

increase kMaxValueLength to 8192 (#27521)

xieofxie committed 18 days ago

Verified 3d1365e2

Account for ORT_NO_EXCEPTIONS builds in Lora test (#27537)

yuslepukhin committed 18 days ago

Verified fdead1ca

Remove some dead code. (#27320)

eserscor committed 18 days ago

Verified eaf8d943

Fix incorrect warning suppression in projects using precompiled headers on Windows (#27039)

skottmckay committed 19 days ago

Verified e7e64dc1

Add "library_path" metadata entry to OrtEpDevice instances for plugin and provider bridge EPs (#27522)

skottmckay committed 19 days ago

Verified c1d8f5c0

[NvTensorRTRTX EP]: Add missing override specifiers to suppress warnings (#27288)

theHamsta committed 19 days ago

Verified 0b2b6d0c

[CXX Lora] Prevent heap OOB from maliciously crafted Lora Adapters. (#27518)

yuslepukhin committed 19 days ago

Verified ef04b10a

Support broadcasting skip shapes in SkipLayerNorm fusion (#27489)

Rishi-Dave committed 19 days ago

Verified 12814f71

Fix for Cuda Graph in pre-compiled path (#27477)

umangb-09 committed 19 days ago

Verified 06bbcd89

[Build] Fix pybind11 vcpkg configuration (#27514)

tianleiwu committed 19 days ago

Verified f99dcca1

Fix -Warray-bounds build error in MLAS on clang 17+ (#27499)

Copilot committed 19 days ago

Verified c7c939ff

Fix double-free in TRT EP custom op domain Release functions (#27471)

Copilot committed 19 days ago

Verified 4a80b0b5

Suppress spurious Array Out of Bounds warnings produced by GCC 14.2 compiler on Linux builds (#27454)

vthaniel committed 19 days ago

Verified 0b9906a3

Split out Linux CUDA Python package builds into separate stages (#27490)

edgchen1 committed 19 days ago

Verified 1acddf2c

Add pre-layer normalization support to attention fusion (#27418)

Rishi-Dave committed 19 days ago

Verified 46126133

Fix O(n²) model load time for TreeEnsemble with categorical feature chains (#27391)

vektah committed 20 days ago

Verified 41f9056c

Remove s_kernel_registry_vitisaiep.reset() in deinitialize_vitisai_ep() (#27295)

zz002 committed 21 days ago

Verified 6e72d319

Reduce allocation size in test Gather_oveflow_check from 4GiB to >2GiB (#27483)

adrianlizarraga committed 22 days ago

Verified 0768f424

[CPU] GridSample operator performance improvement on bilinear interpolation… (#27359)

melkap01-Arm committed 22 days ago

Verified b7a8c245

[EP API] header-only adapter for EP API (#26919)

fs-eire committed 22 days ago

Verified 4fd9ebb8

[MLAS/CPU EP]: Introduce a backend kernel selector config in MLAS (#27136)

hariharans29 committed 22 days ago

Verified 0411d41d

Remove cudaStreamSynchronize from CUDA LLM ops for CUDA graph capture compatibility (#27484)

titaiwangms committed 22 days ago

Verified 479dd393

Improve error message when EPContext node is not assigned to an EP (#27474)

adrianlizarraga committed 22 days ago

Verified 5f94d6c8