-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Pull requests: volcengine/verl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[docs] ppo: add a page for PPO algorithm
#1781
opened May 30, 2025 by
eric-haibin-lin
Loading…
5 of 6 tasks
Stabilize loss calculations by clamping KL divergence values
#1779
opened May 30, 2025 by
syo093c
Loading…
[feat] support custom label for specific devices
#1773
opened May 30, 2025 by
ZDJeffrey
Loading…
3 of 6 tasks
[megatron] moonlight fix per_tensor_generator
#1772
opened May 30, 2025 by
ISEEKYAN
Loading…
6 tasks
[DONOTMERGE][rollout] feat: ChatScheduler requests sglang fully async
#1769
opened May 30, 2025 by
chenhaiq
Loading…
[ray] feat: add timeline option for performance analyse
#1768
opened May 30, 2025 by
chenhaiq
Loading…
[ci][fix] Fix
doc_testci workflow pipeline
#1767
opened May 30, 2025 by
hongpeng-guo
Loading…
3 of 6 tasks
[docs] docs/advance/ppo_lora.rst: Train RL(HF) algorithms with LoRA s…
#1755
opened May 29, 2025 by
thelongestusernameofall
Loading…
[Bugfix] fix OOM issue when resuming in async mode
#1748
opened May 29, 2025 by
llkn-2
Loading…
6 tasks done
[BREAKING][Refactor] Support multi
chat_scheduler to reduce event loop blocking
#1737
opened May 28, 2025 by
llkn-2
Loading…
6 tasks done
[BREAKING] [refactor] Unify async rollout under SGLangRollout, and support sglang==0.4.6.post5
#1717
opened May 27, 2025 by
zyzshishui
Loading…
1 of 6 tasks
[feat][BREAKING] Megatron: Support learning rate scheduler
status: need review
#1701
opened May 26, 2025 by
ETOgaosion
Loading…
6 tasks done
[Feat] accelerate reward_fn execution via parallel ray task
#1693
opened May 26, 2025 by
llkn-2
Loading…
6 tasks done
[vllm] fix: ensure AsyncLLM response_length less equal than max_new_tokens in generation_config.json
#1690
opened May 26, 2025 by
Yangruipis
Loading…
1 of 6 tasks
Fix async SGLang rollout OOM and illegal memory access
#1686
opened May 26, 2025 by
jybsuper
Loading…
6 tasks done
[sglang] Feat: Search Tool Invocation in Multi-Turn RL Training
status: need review
#1682
opened May 25, 2025 by
Lins-01
Loading…
7 tasks done
[BREAKING][rollout] feat: concurrency control and refactor for ChatCompletionScheduler
#1678
opened May 25, 2025 by
dirtyDan0
Loading…
6 tasks done
Synchronize the interface of FunctionCallParser in sglang
#1669
opened May 24, 2025 by
GeLee-Q
Loading…
Previous Next
ProTip!
Updated in the last three days: updated:>2025-05-27.