Skip to content

Pull requests: huggingface/trl

New pull request New
124 Open 2,940 Closed
124 Open 2,940 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

Simplify role handling in prepare_multimodal_messages
#5508 opened Apr 10, 2026 by albertvillanova Member Loading…
1
[TPO] experimental TPO trainer
#5506 opened Apr 10, 2026 by kashif Collaborator Loading…
8 tasks
6
[SSD] Added SSD trainer in experimental
#5505 opened Apr 10, 2026 by kashif Collaborator Loading…
8 tasks
13
Add docs and good defaults for DistillationTrainer
#5500 opened Apr 10, 2026 by cmpatino Collaborator Loading…
2 tasks done
6
feat: add Llama 3 training chat template with generation markers
#5493 opened Apr 9, 2026 by RudrenduPaul Contributor Loading…
4 of 8 tasks
6
Set _tokenizer as trainer attribute
#5489 opened Apr 9, 2026 by albertvillanova Member Loading…
11
Deprecate eos_token config parameter
#5481 opened Apr 9, 2026 by albertvillanova Member Loading…
2
Fix is_liger_kernel_available compatibility with liger-kernel-nightly
#5478 opened Apr 8, 2026 by flofiz Loading…
3 of 6 tasks
1
Support messages with images in prepare_multimodal_messages
#5474 opened Apr 8, 2026 by albertvillanova Member Loading…
12
Fix the tests related to Flash Attention 2
#5473 opened Apr 8, 2026 by YangKai0616 Contributor Loading…
2 tasks
@kashif
6
[docs] Add hardware requirements note to quickstart
#5472 opened Apr 7, 2026 by pqbas Loading…
5 of 8 tasks
1
1
Add Qwen3-VL tool calling support
#5469 opened Apr 7, 2026 by qgallouedec Member Loading…
6
Add GLM-4-MoE tool calling support
#5463 opened Apr 6, 2026 by qgallouedec Member Loading…
7
GOLDTrainer VLM support
#5461 opened Apr 6, 2026 by Strongich Loading…
4 of 8 tasks
24
[docs] Clarify dtype defaults between trf v5 and TRL
#5457 opened Apr 4, 2026 by casinca Contributor Loading…
2 of 4 tasks
1
[AsyncGRPO] Support async tool calls in AsyncRolloutWorker
#5446 opened Apr 3, 2026 by PoilZero Loading…
5 of 8 tasks
6
FIPO loss
#5434 opened Apr 2, 2026 by kdubovikov Contributor Loading…
4 of 8 tasks
3
feat(async-grpo): add sampling parameter parity
#5418 opened Mar 31, 2026 by kdubovikov Contributor Loading…
4 of 8 tasks
Delta weight sync using Xet buckets
#5417 opened Mar 31, 2026 by AmineDiro Member • Draft
8 tasks
3
fix(async-grpo): honor model init dtype
#5416 opened Mar 31, 2026 by kdubovikov Contributor Loading…
3 of 8 tasks
2
Skip redundant forward pass for on-policy vLLM importance sampling
#5413 opened Mar 31, 2026 by GJ98 Loading…
3 of 8 tasks
1
10
add JEPO trainer
#5411 opened Mar 31, 2026 by zbills Loading…
3 of 7 tasks
18
Add log_multimodal param to GRPOConfig and RLOOConfig to control image logging
#5408 opened Mar 30, 2026 by apardyl Contributor Loading…
3 of 8 tasks
4
Add length-normalized sigmoid loss type to DPO trainer
#5406 opened Mar 30, 2026 by BrownianNotion Loading…
5 of 8 tasks
1
7
Add per-sample tool filtering to GRPOTrainer via tools column
#5398 opened Mar 27, 2026 by lailanelkoussy Contributor Loading…
3 tasks done
34
ProTip! Filter pull requests by the default branch with base:main.