Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Feb 13, 2026

• docs: add Multi-Node Training subsection (#4384) (#5091)

  Show description for 7694e0e
  

  nabin2004

  and

  sergiopaniego

  authored
  7694e0eView commit details

  Copy full SHA for 7694e0e

  Browse repository at this point

Commits on Feb 12, 2026

• docs: Add MPO paper (2411.10442) to paper index (#5089)

  Show description for 28fc3f2
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  28fc3f2View commit details

  Copy full SHA for 28fc3f2

  Browse repository at this point

• Validate reward model has 1 num_labels (#5087)

  albertvillanova

  authored
  051b52fView commit details

  Copy full SHA for 051b52f

  Browse repository at this point

• Fix BFD packing for SFT datasets (#5076)

  albertvillanova

  authored
  a558fbaView commit details

  Copy full SHA for a558fba

  Browse repository at this point

• docs: Add PPO paper (1707.06347) to paper index (#5085)

  Show description for 0073db9
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  0073db9View commit details

  Copy full SHA for 0073db9

  Browse repository at this point

• docs: Add T5 packing paper (1910.10683) to paper index (#5084)

  Show description for 29ed9cb
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  29ed9cbView commit details

  Copy full SHA for 29ed9cb

  Browse repository at this point

• docs: Add PRM paper (2211.14275) to paper index (#5083)

  Show description for d0e06fc
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  d0e06fcView commit details

  Copy full SHA for d0e06fc

  Browse repository at this point

• docs: Add GKD paper (2306.13649) to paper index (#5082)

  Show description for ee979a9
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  ee979a9View commit details

  Copy full SHA for ee979a9

  Browse repository at this point

• docs: Add CPO paper (2401.08417) to paper index (#5081)

  Show description for ff84817
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  ff84817View commit details

  Copy full SHA for ff84817

  Browse repository at this point

• docs: Add ORPO paper (2403.07691) to paper index (#5080)

  Show description for fe890df
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  fe890dfView commit details

  Copy full SHA for fe890df

  Browse repository at this point

• docs: Add TR-DPO paper (2404.09656) to paper index (#5078)

  behroozazarkhalili

  authored
  c88f8c5View commit details

  Copy full SHA for c88f8c5

  Browse repository at this point

Commits on Feb 11, 2026

• [SFT] Fix high vRAM consumption during eval with liger kernel (#5069)

  Show description for 0562c3f
  

  3 peopleauthored
  0562c3fView commit details

  Copy full SHA for 0562c3f

  Browse repository at this point

• Fix CI ValueError: Pointer argument (at 0) cannot be accessed from Triton (cpu tensor?) (#5074)

  Show description for 6b38db6
  

  albertvillanova

  and

  qgallouedec

  authored
  6b38db6View commit details

  Copy full SHA for 6b38db6

  Browse repository at this point

• Update model from SequenceClassification to CausalLM in `RewardTrainer` tests (#5060)

  qgallouedec

  authored
  060fbfeView commit details

  Copy full SHA for 060fbfe

  Browse repository at this point

• Fix logging warning suppression for transformers 4.56.2 (#5077)

  albertvillanova

  authored
  0933b7fView commit details

  Copy full SHA for 0933b7f

  Browse repository at this point

• fix: Set `num_labels` to 1 in causal model initialization for RewardTrainer (#5066)

  qgallouedec

  authored
  a07fb82View commit details

  Copy full SHA for a07fb82

  Browse repository at this point

• Fix GRPO multi-turn training with liger kernels (#4975)

  albertvillanova

  authored
  29fe682View commit details

  Copy full SHA for 29fe682

  Browse repository at this point

• fix: Use `launch_args` for all trainers (#5059)

  qgallouedec

  authored
  68399dfView commit details

  Copy full SHA for 68399df

  Browse repository at this point

• Fix logging warning suppression with scoped override for seq-clf head key (#5058)

  qgallouedec

  authored
  d1b066fView commit details

  Copy full SHA for d1b066f

  Browse repository at this point

• docs: Add SimPO paper (2405.14734) to paper index (#5071)

  Show description for 0c3d33b
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  0c3d33bView commit details

  Copy full SHA for 0c3d33b

  Browse repository at this point

• docs: Add RPO paper (2405.16436) to paper index (#5070)

  Show description for f23e3a7
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  f23e3a7View commit details

  Copy full SHA for f23e3a7

  Browse repository at this point

• docs: Add XPO (2405.21046) to Paper Index (#5068)

  Show description for e46005c
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  e46005cView commit details

  Copy full SHA for e46005c

  Browse repository at this point

• docs: Add REINFORCE++ (2501.03262) to Paper Index (#5062)

  Show description for 6d9bba1
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  6d9bba1View commit details

  Copy full SHA for 6d9bba1

  Browse repository at this point

• docs: Add INTELLECT-2 (2505.07291) to Paper Index (#5061)

  Show description for b992e92
  

  behroozazarkhalili

  and

  sergiopaniego

  authored
  b992e92View commit details

  Copy full SHA for b992e92

  Browse repository at this point

• docs: add DeepSeek-R1 training dynamics and GRPO example (#5053)

  Show description for c985dba
  

  JenWei0312

  and

  sergiopaniego

  authored
  c985dbaView commit details

  Copy full SHA for c985dba

  Browse repository at this point

Commits on Feb 10, 2026

• Remove deprecated mergekit_utils moved to experimental (#5057)

  albertvillanova

  authored
  d934eb7View commit details

  Copy full SHA for d934eb7

  Browse repository at this point

• Remove duplicated tests for SFT and add gradient checkpointing tests (#5054)

  qgallouedec

  authored
  991fd07View commit details

  Copy full SHA for 991fd07

  Browse repository at this point

• Remove deprecated classes moved to experimental (#5044)

  albertvillanova

  authored
  d42b23fView commit details

  Copy full SHA for d42b23f

  Browse repository at this point

• Remove deprecated RLOOConfig.max_prompt_length (#5056)

  albertvillanova

  authored
  e1a84cfView commit details

  Copy full SHA for e1a84cf

  Browse repository at this point

• Remove deprecated XPO after moved to experimental (#5055)

  albertvillanova

  authored
  fc56037View commit details

  Copy full SHA for fc56037

  Browse repository at this point

• Remove deprecated PRM after moved to experimental (#5052)

  albertvillanova

  authored
  13bd37eView commit details

  Copy full SHA for 13bd37e

  Browse repository at this point

• Remove deprecated PPO after moved to experimental (#5051)

  albertvillanova

  authored
  0aea314View commit details

  Copy full SHA for 0aea314

  Browse repository at this point

• Remove deprecated ORPO after moved to experimental (#5050)

  albertvillanova

  authored
  d705ac4View commit details

  Copy full SHA for d705ac4

  Browse repository at this point

• Upgrade GitHub Actions to latest versions (#4893)

  Show description for b393c6b
  

  salmanmkc

  and

  qgallouedec

  authored
  b393c6bView commit details

  Copy full SHA for b393c6b

  Browse repository at this point

• Remove deprecated Judges after moved to experimental (#5048)

  albertvillanova

  authored
  ce2ea74View commit details

  Copy full SHA for ce2ea74

  Browse repository at this point