close
Skip to content

Pull requests: huggingface/trl

New pull request New
130 Open 3,102 Closed
130 Open 3,102 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

fix: prevent 5 GB+ CUDA memory leak in activation offloading by syncing streams and clear stashes in OffloadActivations.__exit__
#5700 opened May 4, 2026 by butterwecksolutions Contributor Loading…
2 of 6 tasks
@kashif
5
Reject parallelism_config with cp_size>1 or sp_size>1 in GRPO/RLOO
#5699 opened May 4, 2026 by kashif Collaborator Loading…
8 tasks
2
Add MFU helpers
#5698 opened May 4, 2026 by AmineDiro Member Loading…
1 task done
6
[experimental] Add OpenReward Standard environment adapter
#5696 opened May 3, 2026 by adithya-s-k Collaborator Loading…
3 of 6 tasks
19
Final logits softcapping support for async GRPO Trainer
#5691 opened May 2, 2026 by mlarnouhet Loading…
3 of 8 tasks
1
@AmineDiro
2
Enable chunked NLL loss with VLM in SFT
#5684 opened Apr 29, 2026 by qgallouedec Member Loading…
6
Reduce inconsistency across trainer test files
#5678 opened Apr 29, 2026 by qgallouedec Member Loading…
5
Enable chunked NLL loss with PEFT in SFT
#5676 opened Apr 28, 2026 by qgallouedec Member Loading…
10
feat(vllm-serve): add --reasoning-parser and --reasoning-config flags
#5672 opened Apr 28, 2026 by kfirah-create Loading…
2 of 8 tasks
DeepSeek v4
#5641 opened Apr 25, 2026 by qgallouedec Member • Draft
8 tasks
Align tiny-Glm4MoeForCausalLM with GLM-4.5 reference config
#5638 opened Apr 24, 2026 by qgallouedec Member Loading…
8 tasks
6
Refactor tiny-model generation scripts
#5637 opened Apr 24, 2026 by qgallouedec Member Loading…
23
feat: Add generation_kwargs support to LogCompletionsCallback and Wea…
#5625 opened Apr 22, 2026 by LhaseParth2610 Loading…
4 of 8 tasks
4
Add LoRA support for AsyncGRPO
#5610 opened Apr 21, 2026 by jonahsamost Loading…
2 of 8 tasks
10
experimental: Self-Distillation Zero
#5609 opened Apr 20, 2026 by LeonEricsson Collaborator Loading…
1 of 8 tasks
3
support prefetch/prefetch_depth for async GRPO for ~5% speedups
#5602 opened Apr 20, 2026 by winglian Contributor Loading…
1 of 8 tasks
@AmineDiro
3
Fix nested vocab_size for DistillationTrainer and GOLDTrainer
#5592 opened Apr 19, 2026 by Beichen-Ma Loading…
2 of 8 tasks
1
@cmpatino
1
feat: add TargetPO trainer
#5591 opened Apr 18, 2026 by JeanKaddour • Draft
4 of 8 tasks
5
Add training chat template for Qwen3-2507
#5574 opened Apr 16, 2026 by SwayamInSync Contributor Loading…
5
refactor: self distillation trainers (sdpo/sdft/...)
#5573 opened Apr 16, 2026 by LeonEricsson Collaborator Loading…
2 of 8 tasks
19
Fix empty-target self-distillation loss to stay connected to model graph
#5572 opened Apr 16, 2026 by walawalagoose Loading…
3 of 8 tasks
Improve BrowserGym examples for latest OpenEnv version
#5568 opened Apr 16, 2026 by sergiopaniego Member Loading…
8 tasks
7
Revert VLM support in parse_response
#5561 opened Apr 15, 2026 by qgallouedec Member Loading…
7
Accept processor in get_training_chat_template
#5560 opened Apr 15, 2026 by qgallouedec Member Loading…
8
Move experimental example scripts into their trainer folders
#5556 opened Apr 15, 2026 by sergiopaniego Member Loading…
1 of 8 tasks
1
ProTip! Follow long discussions with comments:>50.