Skip to content

Pull requests: huggingface/trl

New pull request New
66 Open 1,541 Closed
66 Open 1,541 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

🧪 testing suport for Qwen3 tiny
#3415 opened May 5, 2025 by shirinyamani Loading…
5 tasks
7
Feature: Implemented DAPO
#3413 opened May 5, 2025 by lavaman131 Loading…
4
Reintroducing step method in ppo_trainer
#3410 opened May 3, 2025 by jskaf34 • Draft
2 of 5 tasks
1
fix setup chat format
#3404 opened May 2, 2025 by qgallouedec • Draft
5 tasks
[DPO] Truncation leading to zero'd out samples
#3398 opened May 1, 2025 by LeonEricsson Loading…
2 of 5 tasks
3
Fix GRPO/DAPO/Dr.GRPO documentation: formula corrections and KL divergence clarification
#3395 opened Apr 30, 2025 by JenWei0312 Loading…
1 of 5 tasks
9
Reintroduce generate method for PPOTrainer
#3374 opened Apr 27, 2025 by CloseChoice Loading…
4 tasks done
4
An Unified Example Format Checker
#3373 opened Apr 27, 2025 by innerNULL Loading…
1 of 5 tasks
2
add support for reward func using nn.Module in GRPOTrainer
#3372 opened Apr 27, 2025 by Tavish9 Loading…
1 of 5 tasks
4
[Feat] Suppport SGLang as rollout engine of GRPO trainer
#3370 opened Apr 27, 2025 by ryang-max Loading…
2 of 8 tasks
17
Environments
#3367 opened Apr 26, 2025 by August-murr Loading…
3
[GRPO] adds experimental support for the SSR replay buffer
#3325 opened Apr 18, 2025 by edbeeching • Draft
[vllm] support base_url parameter for vLLM client initialization
#3324 opened Apr 18, 2025 by re-imagined Loading…
13
Allow for saving the PPOTrainer value model (critic model)
#3308 opened Apr 16, 2025 by AMindToThink Loading…
PPO value_model can't be None, so it shouldn't be Optional
#3300 opened Apr 15, 2025 by AMindToThink Loading…
5
Modified GRPOTrainer to accumulate gradient within a single training batch
#3288 opened Apr 13, 2025 by jarrelscy Loading…
3 of 5 tasks
6
add vllm support for token ids as input
#3280 opened Apr 11, 2025 by wybryan Loading…
1
🦙 Llama 4
#3267 opened Apr 9, 2025 by qgallouedec • Draft
5 tasks
[NOT MEANT TO BE MERGED] Log correct/incorrect lengths
#3263 opened Apr 8, 2025 by qgallouedec • Draft
[SFT] support for ring_attn in SFTTrainer
#3262 opened Apr 8, 2025 by kashif Loading…
5 tasks
2
Add a raw generate API to the vLLM server
#3227 opened Apr 3, 2025 by wilrop Loading…
5 tasks
4
Support iterable datasets in GRPO
#3226 opened Apr 3, 2025 by wilrop Loading…
5 tasks
1
8
feat(trainer): Support multi-role & consecutive turns in DataCollatorForCompletionOnlyLM (#3223)
#3224 opened Apr 3, 2025 by Kirili4ik Loading…
4 tasks done
1
Adding sampling parameters for vllm generation
#3210 opened Apr 2, 2025 by shaipranesh2 Loading…
6
Support for Models With Pre-Finetuned LoRA Adapters in GRPO: Add use_peft_as_reference Flag
#3196 opened Mar 31, 2025 by LoganVegnaSHOP Loading…
5 tasks done
1
7
ProTip! Adding no:label will show everything without a label.