Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Dec 2, 2025

• Merge branch 'huggingface:main' into main

  ldqvinh

  authored
  108fac4View commit details

  Copy full SHA for 108fac4

  Browse repository at this point

• Remove deprecations for 0.26 release (#4607)

  albertvillanova

  authored
  de343cdView commit details

  Copy full SHA for de343cd

  Browse repository at this point

Commits on Dec 1, 2025

• Silence experimental warnings when imported in the stable (#4606)

  qgallouedec

  authored
  07b4a84View commit details

  Copy full SHA for 07b4a84

  Browse repository at this point

• Update How-to guides (#4604)

  qgallouedec

  authored
  c55ef4bView commit details

  Copy full SHA for c55ef4b

  Browse repository at this point

• Raise FutureWarning for classes moved to experimental (#4605)

  albertvillanova

  authored
  c686d7dView commit details

  Copy full SHA for c686d7d

  Browse repository at this point

• docs: Expand speeding up training guide with acceleration methods (#4428)

  Show description for c7d172b
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  c7d172bView commit details

  Copy full SHA for c7d172b

  Browse repository at this point

• docs: Expand training customization examples (#4427)

  Show description for f1dfef0
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  f1dfef0View commit details

  Copy full SHA for f1dfef0

  Browse repository at this point

Commits on Nov 30, 2025

• [GRPO] Sequence-level TIS & MIS (#4530)

  LeonEricsson

  authored
  eb76389View commit details

  Copy full SHA for eb76389

  Browse repository at this point

Commits on Nov 28, 2025

• docs: Add Beyond the 80/20 Rule (2506.01939) to Paper Index (#4580)

  Show description for 0726977
  

  xuanduy04

  and

  qgallouedec

  authored
  0726977View commit details

  Copy full SHA for 0726977

  Browse repository at this point

• Revert "Hotfix CI with dev dependencies: xfail test_prepare_inputs_for_generation" (#4587)

  Show description for 9731d08
  

  albertvillanova

  and

  qgallouedec

  authored
  9731d08View commit details

  Copy full SHA for 9731d08

  Browse repository at this point

• Fix 'generation_config' AttributeError (#4596)

  albertvillanova

  authored
  84a0bbcView commit details

  Copy full SHA for 84a0bbc

  Browse repository at this point

• Remove module-level imports of extra deps in experimental.judges (#4598)

  albertvillanova

  authored
  f67c3f2View commit details

  Copy full SHA for f67c3f2

  Browse repository at this point

Commits on Nov 27, 2025

• Add missing require_bitsandbytes marker to CI tests (#4586)

  albertvillanova

  authored
  cb5fdf9View commit details

  Copy full SHA for cb5fdf9

  Browse repository at this point

Commits on Nov 26, 2025

• fix: use shift_labels for metrics when using CP or SP (#4579)

  Show description for 4a3b584
  

  jue-jue-zi

  and

  kashif

  authored
  4a3b584View commit details

  Copy full SHA for 4a3b584

  Browse repository at this point

• Revert hotfix Fall back to config.text_config._name_or_path (#4581)

  albertvillanova

  authored
  d2e4315View commit details

  Copy full SHA for d2e4315

  Browse repository at this point

• Move tests for GSPOTokenTrainer to experimental (#4572)

  qgallouedec

  authored
  357e331View commit details

  Copy full SHA for 357e331

  Browse repository at this point

• Move `WinRateCallback` to experimental (#4558)

  Show description for a59f2cf
  

  3 peopleauthored
  a59f2cfView commit details

  Copy full SHA for a59f2cf

  Browse repository at this point

• Fix PPO example (#4556)

  qgallouedec

  authored
  cf431dbView commit details

  Copy full SHA for cf431db

  Browse repository at this point

Commits on Nov 25, 2025

• Fix Replay Buffer docs. (#4574)

  pramodith

  authored
  cac9f1dView commit details

  Copy full SHA for cac9f1d

  Browse repository at this point

• Add `shuffle_dataset` option to `SFTTrainer` (#4564)

  qgallouedec

  authored
  547d924View commit details

  Copy full SHA for 547d924

  Browse repository at this point

• Fix typo in GRPO description in README (#4573)

  iliasmerigh

  authored
  b01f8caView commit details

  Copy full SHA for b01f8ca

  Browse repository at this point

• Fix vLLM sleep mode: add collective RPC call to reload weights in vLLM wake-up process (#4571)

  Show description for 7856d3b
  

  3 peopleauthored
  7856d3bView commit details

  Copy full SHA for 7856d3b

  Browse repository at this point

• Reasoning reward (#4563)

  Show description for 64d089e
  

  lewtun

  and

  qgallouedec

  authored
  64d089eView commit details

  Copy full SHA for 64d089e

  Browse repository at this point

• Remove Online DPO from stable trainers section in documentation

  qgallouedec

  committed
  3b7d0e4View commit details

  Copy full SHA for 3b7d0e4

  Browse repository at this point

Commits on Nov 24, 2025

• Reorder documentation TOC to surface key trainer sections (#4565)

  qgallouedec

  authored
  6f3a452View commit details

  Copy full SHA for 6f3a452

  Browse repository at this point

• docs: Rewrite PEFT integration guide with comprehensive examples (#4421)

  Show description for 46af266
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  46af266View commit details

  Copy full SHA for 46af266

  Browse repository at this point

• Add `num_generations_eval` parameter for efficient evaluation (#4458)

  Show description for db4f6e5
  

  mingxuetian

  and

  qgallouedec

  authored
  db4f6e5View commit details

  Copy full SHA for db4f6e5

  Browse repository at this point

• Move OnlineDPOTrainer to experimental module (#4473)

  Show description for 07f3c95
  

  behroozazarkhalili

  and

  qgallouedec

  authored
  07f3c95View commit details

  Copy full SHA for 07f3c95

  Browse repository at this point

Commits on Nov 22, 2025

• [SFT] Log mean token accuracy from Liger kernel (#4302)

  Show description for 4cb1a25
  

  kashif

  and

  qgallouedec

  authored
  4cb1a25View commit details

  Copy full SHA for 4cb1a25

  Browse repository at this point

• docs: add KTO (2402.01306) to Paper Index + link ref to KTOTrainer (#4440)

  Show description for 468b9d4
  

  SSusantAchary

  and

  qgallouedec

  authored
  468b9d4View commit details

  Copy full SHA for 468b9d4

  Browse repository at this point

• Move PRMTrainer to trl.experimental.prm (#4483)

  Show description for 9bc6206
  

  behroozazarkhalili

  and

  qgallouedec

  authored
  9bc6206View commit details

  Copy full SHA for 9bc6206

  Browse repository at this point

Commits on Nov 21, 2025

• Update OpenEnv guide with new notebook (#4555)

  sergiopaniego

  authored
  f7ac974View commit details

  Copy full SHA for f7ac974

  Browse repository at this point

• Add GRPO Wordle OpenEnv Colab (#4542)

  sergiopaniego

  authored
  c0de042View commit details

  Copy full SHA for c0de042

  Browse repository at this point

• [ORPO] Move ORPOTrainer to experimental (#4480)

  behroozazarkhalili

  authored
  9f8ef40View commit details

  Copy full SHA for 9f8ef40

  Browse repository at this point

• fix+docs: `device_map=None` for DeepSpeed and add ZeRO paper (1910.02054) to Paper Index (#4551)

  JenWei0312

  authored
  3bb5d76View commit details

  Copy full SHA for 3bb5d76

  Browse repository at this point