Skip to content

Pull requests: huggingface/trl

New pull request New
81 Open 2,095 Closed
81 Open 2,095 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

GRPO: ScaleRL -> Support casting LM Head to FP32
#4303 opened Oct 18, 2025 by pramodith Loading…
4 of 5 tasks
2
[SFT] Log mean token accuracy from Liger kernel
#4302 opened Oct 18, 2025 by kashif Loading…
5 tasks
1
⚰️ Remove deprecated
#4301 opened Oct 18, 2025 by qgallouedec Loading…
5 tasks
1
Tool call
#4300 opened Oct 18, 2025 by qgallouedec • Draft
5 tasks
Add CISPO loss option and documentation
#4298 opened Oct 16, 2025 by gustavorubim Loading…
Replace unittest skipTest from transformers with pytest.skip
#4297 opened Oct 16, 2025 by albertvillanova Loading…
7
switch to sleep level=2 and split wake-ups in GRPO and RLOO trainers
#4296 opened Oct 16, 2025 by xxrjun Loading…
1 of 5 tasks
4
feat: Add Multi-Token Prediction (MTP) support to SFTTrainer
#4290 opened Oct 15, 2025 by KLGR123 Loading…
fix CI issue for vlm_gemma_3n model
#4278 opened Oct 15, 2025 by kaixuanliu Loading…
1
[SFT] add support for unified conversion logic for both images and videos
#4264 opened Oct 13, 2025 by kashif Loading…
3
Add Humanline
#4261 opened Oct 13, 2025 by Muennighoff Loading…
2
Remove FSDP1 support: use FSDP2 exclusively
#4260 opened Oct 11, 2025 by behroozazarkhalili Loading…
1
Remove BestOfNSampler class
#4259 opened Oct 11, 2025 by behroozazarkhalili Loading…
1
Fix DPO Trainer Bug For Qwen2-VL (Issue 2660)
#4257 opened Oct 11, 2025 by FabianSchuetze Loading…
1 of 3 tasks
1
Online-dpo-ben
#4252 opened Oct 10, 2025 by burtenshaw • Draft
5 tasks
[Activation-checkpointing] add tensor dedup and param offloading
#4247 opened Oct 10, 2025 by kashif Loading…
13
Added SFT LoRA notebook
#4244 opened Oct 9, 2025 by sergiopaniego Loading…
5 tasks
12
[Utils] fix DataCollatorForChatML
#4231 opened Oct 8, 2025 by kashif • Draft
5
Add support for Python 3.14
#4225 opened Oct 8, 2025 by albertvillanova Loading…
4
Update max_length explanation for VLM trainers
#4220 opened Oct 7, 2025 by sergiopaniego Loading…
5 tasks
3
Add trust_remote_code to GRPOConfig
#4186 opened Oct 1, 2025 by muupan Loading…
3 of 4 tasks
1
6
🐍 Drop Python 3.9
#4183 opened Sep 30, 2025 by qgallouedec Loading…
11
🧺 [5/N] Refactor _generate in GRPO/RLOO: Insert images in the prompt
#4155 opened Sep 26, 2025 by qgallouedec Loading…
1
Fix GKD Liger memory spike
#4140 opened Sep 24, 2025 by qgallouedec Loading…
@kashif
2
update guided decoding param to structured outputs
#4117 opened Sep 22, 2025 by jiqing-feng Loading…
5
ProTip! Updated in the last three days: updated:>2025-10-15.