Skip to content

Pull requests: huggingface/trl

New pull request New
65 Open 1,314 Closed
65 Open 1,314 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

Liger GRPO support
#2926 opened Feb 21, 2025 by SalmanMohammadi • Draft
4 tasks
2
chore(typo): fix typo in TextEnvironment init param, should be max_to…
#2921 opened Feb 21, 2025 by shenxiangzhuang Loading…
1 of 5 tasks
updated DPO default values for alpha and tau
#2918 opened Feb 20, 2025 by Ishan-Kumar2 Loading…
4 of 5 tasks
1
Update kto_config.py
#2912 opened Feb 20, 2025 by sileod Loading…
5 tasks
Remove CUDA synchronization in mean_token_accuracy
#2902 opened Feb 19, 2025 by cyyever Loading…
1 task done
5
GRPOTrainer adds support for OpenAI API-compatible servers to models that generate samples
#2901 opened Feb 19, 2025 by ZYM66 • Draft
2 of 5 tasks
5
parameterize enable_prefix_caching
#2900 opened Feb 19, 2025 by ji-huazhong Loading…
1 of 5 tasks
1
3
Fast dataset truncate in SFTTrainer
#2898 opened Feb 18, 2025 by mariosasko Loading…
1 of 5 tasks
[GRPO] Log completions at every step
#2893 opened Feb 18, 2025 by lidiya-co Loading…
1 of 5 tasks
5
[Discussion] Agentic Framework Based on VLLM and E2B for RL
#2880 opened Feb 17, 2025 by August-murr • Draft
1
[GRPO] Pass wrapped model to unwrap_model_for_generation for DeepSpeed Stage-3 compatibility
#2871 opened Feb 15, 2025 by kiddj Loading…
1 of 5 tasks
2
[GRPO] Reduce steps where loss starts to remain at 0, accelerate training
#2869 opened Feb 15, 2025 by zhangsheng377 Loading…
4
BCOTrainer version upgrade fixes
#2867 opened Feb 15, 2025 by claralp Loading…
3 of 5 tasks
4
Using model_wrapped can improve the generation speed by approximately 10 times on a single GPU.
#2859 opened Feb 14, 2025 by dignfei Loading…
2
Add the metrics completion_length_max and completion_length_min
#2857 opened Feb 14, 2025 by dignfei Loading…
1
grpo_error 😴 stale No update from the author, will be closed soon
#2841 opened Feb 12, 2025 by huihuiustc Loading…
3 of 5 tasks
5
Add GRPO Trainer support for third-party accelerators
#2836 opened Feb 12, 2025 by ji-huazhong Loading…
1 of 5 tasks
13
🔁 Add retry mechanism for GRPO training with configurable reward thre…
#2823 opened Feb 10, 2025 by mandeep511 Loading…
3 of 5 tasks
4
[WIP] [Liger] Liger KTO support
#2812 opened Feb 10, 2025 by vaibhavjindal • Draft
5 tasks
1
GRPO Environments for custom multi-step rollouts (vLLM-only)
#2810 opened Feb 9, 2025 by willccbb Loading…
5 tasks done
34
[draft] Use vLLM in LogCompletionsCallback
#2797 opened Feb 7, 2025 by tchang1997 • Draft
2 of 4 tasks
Remote GRPO ref model
#2763 opened Feb 4, 2025 by edbeeching • Draft
8
allow ref_model to be set in trainer to interface parity with other trainers
#2746 opened Feb 3, 2025 by winglian • Draft
5 tasks
WIP: RLOOV2
#2724 opened Jan 31, 2025 by mnoukhov • Draft
3 tasks
1
🔧 Optimize GRPO VRAM Usage
#2669 opened Jan 27, 2025 by andyl98 Loading…
2 of 5 tasks
38
ProTip! Type g i on any issue or pull request to go back to the issue listing page.