Pull Requests huggingface/peft

detect if torch.distributed is available

#2963 by vladmandic was merged 2025-12-17 11:35

FIX: Inject from state dict into compiled model

#2962 by BenjaminBossan was merged 2025-12-17 11:56

Fix/adadora issues

#2959 by OrMullerHahitti was closed 2025-12-17 13:02

CI For transformers main tests, clear disk space

#2956 by BenjaminBossan was merged 2025-12-18 10:52

TST: Remove unnecessary prefix tuning dtype test

#2955 by BenjaminBossan was merged 2025-12-15 12:06

Add cartridges to PEFT

#2953 by kashif was merged 2026-01-07 16:16

FIX Migrate method comparison space to Gradio 6

#2947 by BenjaminBossan was merged 2025-12-10 12:41

TST Remove unnecessary PREFIXES constant

#2942 by BenjaminBossan was merged 2025-12-10 13:53

Test cleaning pytest caches

#2938 by githubnemo was merged 2025-12-08 13:59

FIX: Some GPU tests failing due to transformers v5

#2937 by BenjaminBossan was merged 2025-12-04 15:22

FIX Don't set requires_grad on original module

#2936 by BenjaminBossan was merged 2025-12-04 15:22

Fix EETQ GPU Docker image build

#2935 by githubnemo was merged 2025-12-03 15:38

FIX Transformers v5 fixes

#2934 by BenjaminBossan was merged 2025-12-03 16:14

Testing Transformers v5 RC

#2933 by BenjaminBossan was closed 2025-12-02 14:22

MetaMath: Add forgetting metric

#2925 by BenjaminBossan was merged 2025-12-03 14:38

Claude/peft adapter architecture 012 eo c nt l axd cdz q3 r9m rcfh

#2924 by TamerSoliman was closed 2025-11-22 18:15

Deal with weight tying in transformers >=5

#2922 by githubnemo was merged 2025-11-20 16:04

FIX Beam search w/ mixed adapter batches & encoder

#2921 by BenjaminBossan was merged 2025-11-20 15:21

FIX Move further models to safetensors

#2920 by BenjaminBossan was merged 2025-11-20 15:20

Added an experimental notebook demonstrating usage of `PeftMixedModel`

#2919 by Tfloow was closed 2025-11-23 10:12

FIX Bug when merging negatively weighted adapters

#2918 by BenjaminBossan was merged 2025-11-20 15:21

[WIP]: Clean auto_gptq and auto_awq

#2917 by LRL2-ModelCloud was closed 2025-12-11 15:22

FIX Load quantized weights with PEFT mixed model

#2915 by BenjaminBossan was merged 2025-11-20 15:21

TST FIX: Issue with pickle models and caching

#2913 by BenjaminBossan was merged 2025-11-19 15:36

Fix caching for LoRA parametrizations on nn.Parameter

#2912 by jonnyli1125 was merged 2025-11-21 13:19

Add Intruder Dimension Mitigation for LoRA Adapters

#2911 by ved1beta was closed 2026-01-07 15:05

Bump version to 0.18.1.dev0 after release

#2910 by BenjaminBossan was merged 2025-11-20 15:20

Add Marian to author list

#2909 by BenjaminBossan was merged 2025-11-20 15:20

Release: v0.18.0

#2908 by BenjaminBossan was merged 2025-11-13 11:03

Resolve #2431: Remove macos-13 from tests

#2906 by githubnemo was merged 2025-11-13 15:46