-
Notifications
You must be signed in to change notification settings - Fork 736
Pull requests: PaddlePaddle/FastDeploy
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[Optimization] Auto set num_max_dispatch_tokens_per_rank
#7237
opened Apr 8, 2026 by
RichardWooSJTU
Loading…
2 of 5 tasks
[Cherry-Pick][Speculative Decoding] Remove arctic_inference deps (#7231)
#7236
opened Apr 8, 2026 by
Deleter-D
Loading…
2 of 5 tasks
[Cherry-Pick][Speculative Decoding] Remove arctic_inference deps (#7231)
#7235
opened Apr 8, 2026 by
Deleter-D
Loading…
2 of 5 tasks
[Cherry-Pick][Optimization] Enable text-only deployment for multimodal models
contributor
External developers
#7234
opened Apr 8, 2026 by
K11OntheBoat
Loading…
5 tasks
[Cherry-Pick][Optimization] Enable text-only deployment for multimodal models(#7183)
#7233
opened Apr 8, 2026 by
EmmonsCurse
Loading…
5 tasks
[Speculative Decoding] Remove arctic_inference deps
#7231
opened Apr 8, 2026 by
Deleter-D
Loading…
2 of 5 tasks
[CI] Support multi-Python version build (3.10/3.11/3.12)
#7230
opened Apr 7, 2026 by
EmmonsCurse
Loading…
5 tasks done
⚡ Bolt: pass std::string by const reference to avoid unnecessary copies in RDMA ops
contributor
External developers
#7229
opened Apr 7, 2026 by
google-labs-jules
bot
Loading…
[Cherry-pick][Optimization] enable trtllm_all_reduce fusion kernel in glm model
#7228
opened Apr 7, 2026 by
BingooYang
Loading…
5 tasks done
[Cherry-Pick][Feature]distinguish whl version(#7204)
#7224
opened Apr 7, 2026 by
EmmonsCurse
Loading…
5 tasks done
[Cherry-pick][Optimization] enable trtllm_all_reduce fusion kernel in glm model
#7219
opened Apr 7, 2026 by
BingooYang
Loading…
5 tasks done
[RL] support moe-topk use topk_reduce_func
#7218
opened Apr 7, 2026 by
zoooo0820
Loading…
2 of 5 tasks
[Cherry-Pick][RL] cherry-pick #7218 support moe-topk use topk_reduce_func
#7217
opened Apr 7, 2026 by
zoooo0820
Loading…
5 tasks
[Cherry-Pick][BugFix] Fix batch_size derivation and relax shape check…
#7216
opened Apr 7, 2026 by
xiaoxiaohehe001
Loading…
5 tasks
[Optimization] Use triton qk_norm both in Prefill and Decode.
cherry-pick: release/2.5
cherry-pick: release/2.6
contributor
External developers
#7213
opened Apr 7, 2026 by
K11OntheBoat
Loading…
5 tasks
[Cherry-Pick][BugFix] Fix batch_size derivation and relax shape check…#7210
#7212
opened Apr 7, 2026 by
xiaoxiaohehe001
Loading…
5 tasks
[BugFix] Fix batch_size derivation and relax shape checks in SM90 flash_mask_attn
#7210
opened Apr 7, 2026 by
xiaoxiaohehe001
Loading…
5 tasks
Previous Next
ProTip!
Exclude everything labeled
bug with -label:bug.