拉取请求: microsoft/onnxruntime
Author
Label
项目
里程碑
Reviews
Assignee
Sort
拉取请求 list
Pin Ruby for Objective-C API docs
#32322
opened Aug 29, 2026 by
Gopalakrishnan Nallasamy (GopalakrishnanN)
Contributor
Loading…
Add packed attention workspace estimates
#32321
opened Aug 29, 2026 by
Ti-Tai Wang (titaiwangms)
Contributor
•
Draft
WebGPU: support local-window PagedAttention with attention sinks
#32320
opened Aug 28, 2026 by
Tianlei Wu (tianleiwu)
Contributor
Loading…
WebGPU: add physical Dawn adapter index selection
#32319
opened Aug 28, 2026 by
Tianlei Wu (tianleiwu)
Contributor
Loading…
[WebGPU] Replace atomic Split-K with a deterministic two-pass reduction
#32318
opened Aug 28, 2026 by
Ananya Anand (4n4ny4)
Contributor
Loading…
[WebGPU] Reject foreign GPU handles in the built-in data transfer
#32317
opened Aug 28, 2026 by
Ananya Anand (4n4ny4)
Contributor
Loading…
[WebGPU] Forward -i provider options to the WebGPU EP in perf_test
#32316
opened Aug 28, 2026 by
Ananya Anand (4n4ny4)
Contributor
Loading…
[WebGPU] Report copy_tensors misuse instead of terminating the process
#32315
opened Aug 28, 2026 by
Ananya Anand (4n4ny4)
Contributor
Loading…
[WebGPU] Select the pooling path by occupancy rather than output size
#32313
opened Aug 28, 2026 by
Ananya Anand (4n4ny4)
Contributor
Loading…
Make workspace input shapes optional-aware
#32312
opened Aug 28, 2026 by
Ti-Tai Wang (titaiwangms)
Contributor
Loading…
Add Rust EPContext data callbacks
#32310
opened Aug 28, 2026 by
Gopalakrishnan Nallasamy (GopalakrishnanN)
Contributor
Loading…
Add Objective-C EPContext data callbacks
#32309
opened Aug 28, 2026 by
Gopalakrishnan Nallasamy (GopalakrishnanN)
Contributor
Loading…
Add JavaScript EPContext data callbacks
#32308
opened Aug 28, 2026 by
Gopalakrishnan Nallasamy (GopalakrishnanN)
Contributor
Loading…
[WebGPU] Pin subgroup size to 32 for subgroup-matrix MatMul/Gemm
#32306
opened Aug 28, 2026 by
Jie Chen (jchen10)
Contributor
Loading…
Add Java EPContext data callbacks
#32305
opened Aug 28, 2026 by
Gopalakrishnan Nallasamy (GopalakrishnanN)
Contributor
Loading…
[WebGPU] Share subgroup matrix MatMul with pointwise Conv
#32304
opened Aug 28, 2026 by
Jiajia Qin (qjia7)
Contributor
•
Draft
Support attention bias with windowed CPU GQA
#32302
opened Aug 28, 2026 by
Tianlei Wu (tianleiwu)
Contributor
Loading…
2 of 3 tasks
Gate CPU FP16 Gemm and MatMul on hardware acceleration
#32301
opened Aug 27, 2026 by
Copilot
AI
Loading…
Update cpuinfo and apply thread-safe deinitialization patch
#32300
opened Aug 27, 2026 by
Vineeth Chelur (crvineeth97)
Contributor
•
Draft
[MLAS] Integrate new optimised SME2 SGEMM and GEMV using common MatMul API
#32296
opened Aug 27, 2026 by
patryk-kaiser-ARM
Contributor
Loading…
[perftest] Add opt-in VitisAI NPU shared allocator for plugin EP I/O
#32295
opened Aug 27, 2026 by
SushmitaThakallapalli1980
Loading…
Enable LayerNormFusion for the WebGPU EP to fix fp16 inference correctness
#32294
opened Aug 27, 2026 by
Lapis0x0 (Lapis0x0)
Loading…
[WebNN] Fix inverted fallback data type support check
#32293
opened Aug 27, 2026 by
Wanming Lin (Honry)
Contributor
Loading…
[CUDA] Extend speculative decode GEMVs to 64 rows
#32289
opened Aug 27, 2026 by
Tianlei Wu (tianleiwu)
Contributor
Loading…
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.