Skip to content

Pull requests: PaddlePaddle/FastDeploy

New pull request New
408 Open 5,060 Closed
408 Open 5,060 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

[RL] support qkrmsnorm use paddle.nn.functional.rms_norm
#6862 opened Mar 16, 2026 by zoooo0820 Loading…
3 of 5 tasks
1
[Optimization][BugFix]Optimize Deepseek networking code
#6861 opened Mar 16, 2026 by chang-wenbin Loading…
5 tasks
1
Dsa
#6860 opened Mar 16, 2026 by zhoutianzi666 Loading…
5 tasks
1
[Cherry-Pick][RL] support qkrmsnorm use paddle.nn.functional.rms_norm
#6859 opened Mar 16, 2026 by zoooo0820 Loading…
3 of 5 tasks
1
update request
#6858 opened Mar 16, 2026 by luukunn Loading…
5 tasks
1
[Feature] Add deepgemm bias epilogue for SM100
#6857 opened Mar 16, 2026 by Wanglongzhi2001 Loading…
5 tasks
1
⚡ Bolt: Optimize RequestMetrics.to_dict() execution time
#6856 opened Mar 15, 2026 by ZeyuChen Loading…
3 tasks done
3
2 5 fa4 support causal
#6854 opened Mar 15, 2026 by ckl117 Loading…
5 tasks
2
⚡ Bolt: Optimize RequestMetrics to_dict serialization
#6853 opened Mar 14, 2026 by ZeyuChen Loading…
4 tasks done
3
[RL][BugFix][Optimization] Support chunked part files loading and fix model path format in IPC snapshot strategy contributor External developers
#6852 opened Mar 14, 2026 by wikilsh Loading…
5 tasks done
2
[Other] Adjust GPUModelRunner to enhance compatibility
#6851 opened Mar 14, 2026 by ming1753 Loading…
5 tasks
2
[RL] Support moe_topk_select using Paddle native operators for training alignment
#6850 opened Mar 14, 2026 by DanielSun11 Loading…
2 of 5 tasks
2
Dev fa4 support causal
#6849 opened Mar 14, 2026 by ckl117 Loading…
5 tasks
2
[Cherry-Pick][RL] Add fused stack-transpose-quant for BlockWiseFP8 MoE weight quantization and swiglu-fp8-quant op for DeepGemmFusedMoE(#6846)
#6848 opened Mar 14, 2026 by zhanghonggeng Loading…
2 of 5 tasks
2
[RL] Add fused swiglu-fp8-quant op support in DeepGemmFusedMoE
#6847 opened Mar 14, 2026 by zhanghonggeng Loading…
2 of 5 tasks
8
[RL] Add fused stack-transpose-quant for BlockWiseFP8 MoE weight quantization and swiglu-fp8-quant op for DeepGemmFusedMoE
#6846 opened Mar 14, 2026 by zhanghonggeng Loading…
2 of 5 tasks
17
[Refactor] support cache transfer manager proxy mode
#6845 opened Mar 14, 2026 by liyonghua0910 Loading…
5 tasks
7
[Performance] Optimize insert_tasks_v1: batch update decode/preempted requests
#6843 opened Mar 13, 2026 by Foriv Loading…
2
⚡ Bolt: Optimize RequestMetrics.to_dict serialization speed
#6841 opened Mar 13, 2026 by ZeyuChen Loading…
5 tasks done
2
[Cherry-pick][Feature] support deepgemm warmup in blackwell
#6838 opened Mar 13, 2026 by lizhenyun01 Loading…
5 tasks
2
[Feature] support deepgemm warmup in blackwell
#6837 opened Mar 13, 2026 by lizhenyun01 Loading…
5 tasks
2
[Optimization][bugfix]optimize dsa model & fix sum bug
#6836 opened Mar 13, 2026 by chang-wenbin Loading…
5 tasks
1
[Optimization][bugfix]optimize dsa model & fix sum bug
#6835 opened Mar 13, 2026 by chang-wenbin Loading…
5 tasks
1
[WIP]add auto split
#6834 opened Mar 13, 2026 by gongweibao Loading…
5 tasks
3
Feature】support flashinfer-cutedsl
#6833 opened Mar 13, 2026 by lizexu123 Loading…
5 tasks
2
ProTip! Follow long discussions with comments:>50.