Pull Requests microsoft/onnxruntime

Add `--build-mode` to packaging pipeline orchestrator (trigger_and_wait_pipelines.py)

#29169 opened 2026-06-19 23:42 by edgchen1

Add reference examples and guidance (especially for multi-device scenarios) for ValidateCompiledModelCompatibilityInfo

#29168 opened 2026-06-19 23:08 by adrastogi

Bump vite and @vitejs/plugin-vue in /js/web/test/e2e/exports/testcases/vite-default dependencies javascript

#29165 opened 2026-06-19 18:05 by dependabot[bot]

Segregate IExecutionProvider optional capabilities into mix-in interfaces

#29087 opened 2026-06-17 02:41 by GopalakrishnanN

Add MLFloat16 QuickGelu CPU kernel for fused fp16 Swish/SiLU

#29080 opened 2026-06-16 17:32 by Copilot

Add 2-bit quantization support to WebGPU GatherBlockQuantized operator ep:WebGPU

#29074 opened 2026-06-16 11:44 by Shivani767

[WebGPU plugin EP] Update release and packaging-related docs

#29059 opened 2026-06-15 21:59 by edgchen1

[web] Fix use-after-free of model buffer with use_ort_model_bytes_for_initializers

#29033 opened 2026-06-12 20:54 by yrliou

Plugin WebGPU EP Branch: Update protobufjs versions

#29031 opened 2026-06-12 18:55 by adrastogi

[CUDA] QMoE support shared experts

#29028 opened 2026-06-12 08:21 by tianleiwu

[ORT] Extend reshape fusion

#29027 opened 2026-06-12 08:00 by Honry

Fix `ToOrtStatus` translation of `SYSTEM`-category `Status` error codes

#29023 opened 2026-06-12 02:24 by edgchen1

fix(react-native e2e): update Android Gradle autolinking and iOS CI runner for react-native 0.86 dependencies javascript

#29019 opened 2026-06-11 23:18 by dependabot[bot]

Fix libatomic linking on toolchains that default to ld --as-needed

#29008 opened 2026-06-11 15:00 by ssam18

Remove unimodule.json so Expo autolinks onnxruntime-react-native

#29005 opened 2026-06-11 13:14 by danielweinmann

Bump torch from 2.7.0 to 2.12.0 in /onnxruntime/python/tools/transformers/models/whisper dependencies python

#28993 opened 2026-06-10 20:51 by dependabot[bot]

Clarify available EP when InferenceSession providers not supplied

#28984 opened 2026-06-10 15:53 by jamesobutler

docs: add PyTorch export guidance and external data file handling (#28763)

#28983 opened 2026-06-10 15:08 by lamjingba-thingom29

S390x workflow using qemu

#28981 opened 2026-06-10 11:01 by AlekseiNikiforovIBM

Fix WebGPU ConvTranspose incorrect output for fp16 models (output spatial dim > 2048)

#28977 opened 2026-06-10 08:05 by shakedm100

Add double support for Cos op on CPU EP

#28975 opened 2026-06-10 07:44 by elwhyjay

Fix CUDA LayerNorm/RMSNorm silent wrong output for rank-reduced scale

#28971 opened 2026-06-10 03:51 by justinchuby

Add Execution Provider conformance test suite

#28968 opened 2026-06-09 23:09 by GopalakrishnanN

Fix SimplifiedLayerNormFusion crash on FP16 models with PrecisionFreeCast (issue #26717)

#28959 opened 2026-06-09 18:50 by Muneerali199

[CUDA EP] Slice: single-memcpy fast path for contiguous subregions

#28902 opened 2026-06-08 20:49 by Copilot

Fix: Add CMake package files to Windows artifact #28468

#28897 opened 2026-06-08 07:18 by Jayashree-mcw

Add Windows ARM64 CUDA plugin package and align CUDA metadata/artifact naming

#28896 opened 2026-06-08 06:38 by tianleiwu

feat: add native CPU kernel for SequenceMap (opset 17)

#28813 opened 2026-06-05 11:34 by Rishi-Dave

[CoreML EP] Add Equal builder

#28807 opened 2026-06-05 08:27 by maxwbuckley

Fix stale output shape metadata in WebGPU external-data E2E model

#28805 opened 2026-06-05 07:11 by Copilot