Skip to content

Pull requests: PaddlePaddle/Paddle

New pull request New
555 Open 56,963 Closed
555 Open 56,963 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

[Dy2St] Fix dy2st recompilation triggered by dictionary order contributor External developers
#76944 opened Dec 17, 2025 by HydrogenSulfate Loading…
6
[Metax] fix device manager release error and adjust the mode of cudag… contributor External developers
#76943 opened Dec 17, 2025 by zhang-chenyi Loading…
2
[Feature] add Blackwell arch support for fused moe
#76942 opened Dec 17, 2025 by Echo-Nie Loading…
2
add_grad share_buffer
#76940 opened Dec 17, 2025 by wanghuancoder Loading…
1
[API Compatibility No.58、349] Sink diag to cpp -part API Compatibility contributor External developers HappyOpenSource 快乐开源活动issue与PR
#76939 opened Dec 16, 2025 by algorithm1832 Loading…
@luotao1 @zhwesky2010
5
[PIR][2/N]【CINN】 Support register Python function on PIR (NativeMetaTensor and lower part)
#76938 opened Dec 16, 2025 by SigureMo Loading…
2
[large tensor] fix slice like x[:, [1,2]]
#76937 opened Dec 16, 2025 by zrr1999 Loading…
1
link rocblas and rocsolver to phi_gpu when magma is enabled
#76936 opened Dec 16, 2025 by HydrogenSulfate Loading…
1
[PIR][1/N]【CINN】 Support register Python function on PIR (IR part) skip-ci: approval
#76935 opened Dec 16, 2025 by SigureMo Loading…
12
[CodeStyle][large tensor] fix some bug where int64 is assigned to int32
#76934 opened Dec 16, 2025 by zrr1999 Loading…
2
Merge fast_rms_norm and fusid_rms_norm into rms_norm
#76930 opened Dec 16, 2025 by zhengshengning Loading…
1
add deepmd-kit to workflow
#76928 opened Dec 16, 2025 by HydrogenSulfate Loading…
1
Cherrypick matmul related devPR into fleety12
#76927 opened Dec 16, 2025 by A-nnonymous Loading…
3
[XPU] fix send/recv hang in dynamic shape XPU
#76925 opened Dec 16, 2025 by lj970926 Loading…
4
Fix hash() when FLAGS_enable_cinn_kernel_cache is false.
#76924 opened Dec 15, 2025 by YuhanXu Loading…
3
Align matmul grad with torch v2.9.1
#76922 opened Dec 15, 2025 by A-nnonymous Loading…
1
[XPU] bind _record_stream contributor External developers
#76921 opened Dec 15, 2025 by ZhouMinhao98 Loading…
10
rename rms_norm to rms_norm_quant
#76920 opened Dec 15, 2025 by zhengshengning Loading…
3
[Type Promotion] Expand Support Datatype
#76919 opened Dec 15, 2025 by Eddie-Wang1120 Loading…
1
Fix PaddleNLP ci test loss for V100.
#76918 opened Dec 15, 2025 by baoqiwen Loading…
8
[API Compatibility No.134、135] align lstm and lstmcell -part API Compatibility contributor External developers HappyOpenSource 快乐开源活动issue与PR
#76915 opened Dec 15, 2025 by Manfredss Loading…
@luotao1 @zhwesky2010
4
[Comp] Add index_elementwise_put_double_grad in eager mode contributor External developers
#76911 opened Dec 15, 2025 by HydrogenSulfate Loading…
2
Remove redundant headers in Paddle CPU
#76910 opened Dec 15, 2025 by YqGe585 Loading…
9
base impl contributor External developers
#76909 opened Dec 15, 2025 by WintersMontagne10335 Loading…
2
[XPU] support 0-size tensor for MaskedSelectGrad XPU
#76905 opened Dec 14, 2025 by cqulilujia Loading…
2
ProTip! no:milestone will show everything without a milestone.