Skip to content

Pull requests: volcengine/verl

New pull request New
236 Open 2,211 Closed
236 Open 2,211 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

[rollout] fix: mlflow consecutive slashes
#4446 opened Dec 8, 2025 by BaiqingL Loading…
9
Routing replay implemented with SGLang and FSDP
#4443 opened Dec 7, 2025 by KawaiiNotHawaii Loading…
7 tasks
10
[rollout, vllm] feat: support KV cache FP8
#4435 opened Dec 5, 2025 by sharonyu-115 Loading…
7 tasks
8
[data] feat: TransferQueue - Add zero-copy serialization support & usage improvement
#4429 opened Dec 5, 2025 by 0oshowero0 Loading…
7 tasks
9
[BREAKING][recipe, ckpt] feat: support parameter sync by checkpoint-engine. only for fully_async mode.
#4427 opened Dec 5, 2025 by zpltys Loading…
5 of 7 tasks
12
feat: allow override for reward_manager_worker in agent loop
#4423 opened Dec 4, 2025 by ryxli Loading…
7 tasks
1
[recipe] fix: Enable optimizer_cpu_offload when training with FP16
#4416 opened Dec 4, 2025 by xichengpro Loading…
7 tasks
1
2
[rollout, sglang] feat: support blockwise fp8 rollout
#4415 opened Dec 4, 2025 by Agoniii Loading…
5 of 7 tasks
4
[trainer] feat: Optimize tokenizer decode calls in validation
#4407 opened Dec 4, 2025 by none0663 Loading…
7 tasks done
3
[model] feat: add qwen25 gspo 3b script on ASCEND NPU
#4405 opened Dec 3, 2025 by zyang6 Loading…
8
[trainer] feat: model engine sft trainer support vlm model
#4403 opened Dec 3, 2025 by wuxibin89 Loading…
2 tasks done
21
fix qwen3vlmoe with Monkey patch to fix a bug in transformers 4.57.x
#4402 opened Dec 3, 2025 by pengyanai Loading…
1 of 7 tasks
6
[trainer] feat: add gpt-oss sglang-megatron example
#4394 opened Dec 3, 2025 by yiakwy-xpu-ml-framework-team Loading…
7 tasks
12
[recipe] fix: fully_async_ppo_trainer hydra search path
#4393 opened Dec 3, 2025 by garrett361 Loading…
5 of 7 tasks
1
fix(vllm): fix typo in async server initialization
#4391 opened Dec 2, 2025 by omkarrr2533 Loading…
10 tasks done
4
[worker] feat: add reusable port
#4377 opened Dec 2, 2025 by burgerkingeater Loading…
2 of 7 tasks
5
[fsdp] feat: integrate PrefixGrouper for GRPO training acceleration
#4368 opened Dec 1, 2025 by kevssim Loading…
3 of 5 tasks
9
multiturn_eval
#4366 opened Dec 1, 2025 by albertimff Loading…
4
[rollout] partial rollout
#4349 opened Nov 29, 2025 by HelloWorldBeginner Loading…
7 tasks
10
[algo] SAPO algo by Qwen
#4345 opened Nov 28, 2025 by BounharAbdelaziz Loading…
5 of 7 tasks
10
[worker] fix: critic/rm support VLM with Ulysses sequence parallel
#4344 opened Nov 28, 2025 by dongju-2 Loading…
4 of 7 tasks
2
[rollout] chore: Add allowed_token_ids into config
#4343 opened Nov 28, 2025 by moehanabi Loading…
4 of 7 tasks
4
[rollout] feat: Support load sampling params from config for vLLMHttpServerBase & SGLangHttpServer
#4342 opened Nov 28, 2025 by moehanabi Loading…
5 of 7 tasks
13
Fix deprecated pynvml dependency warnings
#4339 opened Nov 28, 2025 by Ghazi-raad Loading…
7 tasks
1
4
[WIP][FSDP,Training] feat: Support FSDP2 FP8 training
#4325 opened Nov 27, 2025 by eternally-z Loading…
8 tasks
1
ProTip! Updated in the last three days: updated:>2025-12-05.