-
Notifications
You must be signed in to change notification settings - Fork 2.9k
Pull requests: huggingface/trl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Use the vLLM base image for TRL Docker images
#6687
opened Aug 8, 2026 by
YingqiDuan
Loading…
5 of 8 tasks
[SFT] Treat past_key_values and attentions as optional on the model output
#6686
opened Aug 8, 2026 by
Hakureirm
Loading…
5 of 8 tasks
Use spawn context for vLLM server workers
#6682
opened Aug 8, 2026 by
Sevendogs077
Loading…
5 of 8 tasks
Add an experimental GPG trainer
#6681
opened Aug 8, 2026 by
behroozazarkhalili
Collaborator
Loading…
Document the GRPO reward-function batch layout and group-relative rewards
#6680
opened Aug 8, 2026 by
behroozazarkhalili
Collaborator
Loading…
Pin the TruffleHog scanner image by digest
#6677
opened Aug 7, 2026 by
albertvillanova
Member
Loading…
RFC: pluggable training client for AsyncGRPOTrainer
#6676
opened Aug 7, 2026 by
kashif
Collaborator
Loading…
[GRPO + RLOO] Fix vLLM server-mode generation for multi-turn tool loops
#6673
opened Aug 6, 2026 by
Strongich
Contributor
Loading…
Fix opencode.py proxy trace/log paths after OpenEnv moved them to functions
#6672
opened Aug 6, 2026 by
sergiopaniego
Member
Loading…
1 task done
fix(sft): warn when all labels are masked instead of silent 0.0 loss
#6671
opened Aug 6, 2026 by
navaneethkrishnansuresh
Loading…
4 of 8 tasks
Fix wrapped packing duplicating data when the input table is sliced
#6670
opened Aug 6, 2026 by
SuryanshSS1011
Loading…
4 of 8 tasks
Exclude formatting/template tokens from the per-token KL penalty (#2933)
#6667
opened Aug 5, 2026 by
behroozazarkhalili
Collaborator
Loading…
Restore tokenizer padding after callback generation
#6664
opened Aug 4, 2026 by
RerankerGuo
Loading…
5 of 8 tasks
Restore generation context after errors
#6660
opened Aug 4, 2026 by
RerankerGuo
Loading…
5 of 8 tasks
[DistillationTrainer refactor] Docs: announce the DistillationTrainer promotion in What's New
#6653
opened Aug 3, 2026 by
qgallouedec
Member
•
16/16
Loading…
Allow vLLM weight-update group re-init after a client crash (#3408)
#6651
opened Aug 3, 2026 by
behroozazarkhalili
Collaborator
Loading…
[DistillationTrainer refactor] Support Vision Language Models
#6650
opened Aug 3, 2026 by
qgallouedec
Member
•
14/16
Loading…
[GRPO] Fix entropy bonus normalization inconsistency across loss types
#6648
opened Aug 3, 2026 by
YaseenBashaT
Contributor
Loading…
3 of 8 tasks
[DistillationTrainer refactor] Move the distillation tests to
tests/
#6647
opened Aug 3, 2026 by
qgallouedec
Member
•
10/16
Loading…
[DistillationTrainer refactor] Re-point ServerDistillationTrainer at the stable base
#6645
opened Aug 3, 2026 by
qgallouedec
Member
•
13/16
Loading…
[DistillationTrainer refactor] Update the example script for the stable API
#6644
opened Aug 3, 2026 by
qgallouedec
Member
•
12/16
Loading…
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-07-08.