-
Notifications
You must be signed in to change notification settings - Fork 4.1k
Pull requests: microsoft/onnxruntime
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[Performance] Improve CPU masked attention performance
#29719
opened Jul 14, 2026 by
tianleiwu
Contributor
Loading…
3 of 4 tasks
Bump transformers from 4.52.1 to 5.5.0 in /tools/ci_build/requirements/transformers-test
dependencies
Pull requests that update a dependency file
python
Pull requests that update Python code
#29718
opened Jul 14, 2026 by
dependabot
Bot
Loading…
[onnxruntime-web] Plans to deprecate and remove WebGL and JSEP
#29716
opened Jul 14, 2026 by
edgchen1
Contributor
Loading…
[CUDA] Update cuda arch list for packages of cuda 12.8
release:1.28.0
#29711
opened Jul 14, 2026 by
tianleiwu
Contributor
Loading…
[MLAS][KleidiAI] Add MatMul, Gemm, and Conv FP16 execution paths
#29709
opened Jul 14, 2026 by
JonathanC-ARM
Contributor
Loading…
Fix GroupQueryAttentionFusion crash on ONNX-domain RotaryEmbedding
#29707
opened Jul 14, 2026 by
namgyu-youn
Contributor
Loading…
[CUDA] Fix XQA GroupQueryAttention cudaErrorInvalidValue on Blackwell (sm_120)
release:1.28.0
#29706
opened Jul 14, 2026 by
tianleiwu
Contributor
Loading…
[CUDA] Do not link nvrtc
release:1.28.0
#29705
opened Jul 14, 2026 by
tianleiwu
Contributor
Loading…
Revert "[Plugin EP] Add plugin EP APIs to retrieve ONNX operator sche…
#29702
opened Jul 14, 2026 by
mingyueliuh
Contributor
•
Draft
[Core] Modernize execution synchronization and cancellation
#29700
opened Jul 14, 2026 by
GopalakrishnanN
Contributor
Loading…
Bump transformers from 4.52.1 to 5.5.0 in /onnxruntime/python/tools/transformers/models/llama
dependencies
Pull requests that update a dependency file
python
Pull requests that update Python code
#29696
opened Jul 13, 2026 by
dependabot
Bot
Loading…
Bump transformers from 4.50.0 to 5.5.0 in /onnxruntime/python/tools/transformers/models/stable_diffusion/requirements
dependencies
Pull requests that update a dependency file
python
Pull requests that update Python code
#29695
opened Jul 13, 2026 by
dependabot
Bot
Loading…
[WebGPU] updates webgpu softmax to use online algorithm
#29694
opened Jul 13, 2026 by
prathikr
Contributor
Loading…
Optimize CANN auto padding lookup
#29693
opened Jul 13, 2026 by
GopalakrishnanN
Contributor
Loading…
[CUDA] Gracefully disable MatMulNBits fpA_intB fast path when it cannot serve prefill (narrow-N int4)
#29692
opened Jul 13, 2026 by
justinchuby
Contributor
Loading…
Avoid redundant LUT buffer initialization
#29690
opened Jul 13, 2026 by
GopalakrishnanN
Contributor
Loading…
Size default thread pools from ORT_*_NUM_THREADS environment variables
#29688
opened Jul 13, 2026 by
OscarFree
Loading…
[BUILD] CUDA_QUANT_PREPROCESS off by default and Adjust CI
release:1.28.0
#29687
opened Jul 13, 2026 by
tianleiwu
Contributor
Loading…
Add decode benchmark comparing ONNX Attention CUDA with contrib GQA
#29684
opened Jul 13, 2026 by
namgyu-youn
Contributor
Loading…
[WebGPU EP] Fix unnecessary CPU fallback and improve fallback diagnostic
#29682
opened Jul 13, 2026 by
mingmingtasd
Contributor
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.