generated from fastai/nbdev_template
-
Notifications
You must be signed in to change notification settings - Fork 2.4k
Pull requests: huggingface/trl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix: handle None eval_dataset in example code
#4756
opened Dec 27, 2025 by
ciaoyizhen
Loading…
1 of 4 tasks
perf: avoid output_hidden_states when only last_hidden_state is used
#4755
opened Dec 27, 2025 by
ciaoyizhen
Loading…
2 of 5 tasks
Fix GRPO
scale_rewards type specification to fix __post_init__ validation
#4752
opened Dec 26, 2025 by
apalmas-saifh
Loading…
1 of 5 tasks
Clarify Accelerate usage in SFTTrainer documentation
#4744
opened Dec 23, 2025 by
Likhita-17
Loading…
1 task done
[GRPOTrainer]: Agent Training Supports Async Tool Calls
#4742
opened Dec 23, 2025 by
pramodith
Loading…
5 tasks done
Up to 50% less VRAM during forward with
forward_masked_logits function
#4729
opened Dec 20, 2025 by
qgallouedec
Loading…
Refactor vLLM generation [2/N]: Decouple rollout_func and vLLM functionalities
#4712
opened Dec 17, 2025 by
albertvillanova
•
Draft
Refactor vLLM generation [1/N]: Extract vLLM generation
#4700
opened Dec 16, 2025 by
albertvillanova
Loading…
fix: invalidate ZeRO-3 param coordinator trace in add_hooks
#4693
opened Dec 15, 2025 by
roycho96
Loading…
1 of 5 tasks
feat: DeepSeek V3.2 Off-policy sequence masking
#4689
opened Dec 13, 2025 by
casinca
Loading…
4 of 5 tasks
GKDTrainer: Fix return_outputs in Liger kernel path and update tests
#4688
opened Dec 13, 2025 by
roycho96
Loading…
2 of 5 tasks
CPOTrainer - Incorrect handling of different length chosen/rejected p…
#4639
opened Dec 8, 2025 by
davmels
Loading…
Add cross-tokenizer distillation support for GKD and MiniLLM trainers
#4561
opened Nov 22, 2025 by
sambhavnoobcoder
Loading…
Add PSPO trust region method as alternative to clipping in GRPOTrainer
#4548
opened Nov 19, 2025 by
MCDwyer
Loading…
2 of 5 tasks
Previous Next
ProTip!
Find all pull requests that aren't related to any open issues with -linked:issue.