Skip to content

Pull requests: microsoft/onnxruntime

New pull request New
571 Open 18,803 Closed
571 Open 18,803 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[CPU] Add FP32 flash attention (prefill) and GEMV decode kernel for GroupQueryAttention
#29216 opened Jun 22, 2026 by tianleiwu Contributor • Draft
Document onnxruntime_perf_test usage, input data, and options
#29215 opened Jun 22, 2026 by ArsalanShakil Contributor Loading…
1
List expected types in operator type-mismatch error message
#29214 opened Jun 22, 2026 by ArsalanShakil Contributor Loading…
1
WebGPU: make WebGpuContextFactory::Cleanup exception-safe and idempotent
#29207 opened Jun 22, 2026 by samuel100 Contributor • Draft
1
Fix CUDA/cuDNN DLL preload paths for CUDA 13 consolidated wheel layout
#29202 opened Jun 21, 2026 by tianleiwu Contributor Loading…
2 of 3 tasks
Fix DropQDQ not dropping Q/DQ around opset-22 MaxPool
#29201 opened Jun 21, 2026 by tairenpiao Contributor Loading…
1
1
[Pipeline] Fix Python packaging pipeline failures
#29200 opened Jun 20, 2026 by hariharans29 Member Loading…
10
Fix SimplifiedLayerNorm fusion with node-produced Pow exponent
#29196 opened Jun 20, 2026 by the0cp Contributor Loading…
1
14
Fix NchwcTransformer null deref on node with unresolved schema
#29194 opened Jun 20, 2026 by tairenpiao Contributor Loading…
1
3
[CUDA] Enable CUDA GQA QK-Norm and XQA decode
#29186 opened Jun 20, 2026 by tianleiwu Contributor Loading…
4 tasks done
24
[CUDA] Add sliding-window support to non-quantized XQA decode
#29177 opened Jun 20, 2026 by tianleiwu Contributor Loading…
4 tasks done
18
[CUDA] Fuse MoE router bias into MatMulNBits GEMV
#29170 opened Jun 20, 2026 by tianleiwu Contributor Loading…
26
Add --build-mode to packaging pipeline orchestrator (trigger_and_wait_pipelines.py)
#29169 opened Jun 19, 2026 by edgchen1 Contributor Loading…
Add reference examples and guidance (especially for multi-device scenarios) for ValidateCompiledModelCompatibilityInfo
#29168 opened Jun 19, 2026 by adrastogi Contributor Loading…
4
[CUDA]: Split-K2 QMoE SwiGLU GEMV kernel
#29167 opened Jun 19, 2026 by tianleiwu Contributor Loading…
5
Bump vite and @vitejs/plugin-vue in /js/web/test/e2e/exports/testcases/vite-default dependencies Pull requests that update a dependency file javascript Pull requests that update Javascript code
#29165 opened Jun 19, 2026 by dependabot Bot Loading…
[CPU] Enable pre-packed weights sharing for MatMulNBits
#29163 opened Jun 19, 2026 by derdeljan-msft Contributor Loading…
@derdeljan-msft
3
Harden Android NDK setup against apt mirror outages
#29159 opened Jun 18, 2026 by Copilot AI • Draft
@crvineeth97
Fix over-copy of packed sub-byte tensors in OrtApi::GetValue
#29157 opened Jun 18, 2026 by neilmsft Contributor Loading…
10
Recover Conv/ConvTranspose rank from weight when input shape is unknown
#29149 opened Jun 18, 2026 by fanchenkong1 Contributor Loading…
1
Experimental C++ API update
#29142 opened Jun 17, 2026 by edgchen1 Contributor Loading…
1
Segregate IExecutionProvider optional capabilities into mix-in interfaces
#29087 opened Jun 17, 2026 by GopalakrishnanN Contributor Loading…
9
Avoid small MatMul batch parameter heap allocations
#29085 opened Jun 17, 2026 by GopalakrishnanN Contributor Loading…
4
Fix Shape→Gather→TopK regression: preserve rank-1 single-element index output in data propagation
#29084 opened Jun 16, 2026 by titaiwangms Contributor Loading…
21
Add MLFloat16 QuickGelu CPU kernel for fused fp16 Swish/SiLU
#29080 opened Jun 16, 2026 by Copilot AI • Draft
@tianleiwu
6
ProTip! Exclude everything labeled bug with -label:bug.