Search results

Add support for Gemma4ClippableLinear (Gemma 4 QLoRA fails)

Status: Open.

#3129 In huggingface/peft;

· dentity007 opened on Apr 2, 2026

Nemotron-3-Nano NemotronHTopkRouter

Status: Open.

#3123 In huggingface/peft;

· lucian-student opened on Mar 30, 2026

GraLoRA merge_and_unload with modules_to_save error

Status: Open.

#3114 In huggingface/peft;

· JBurtn opened on Mar 22, 2026

Security: Arbitrary Module Import via Malicious Adapter Config (CWE-94)

Status: Open.

#3085 In huggingface/peft;

· spartan8806 opened on Mar 8, 2026

FSDP + torch.nn.Parameter (MoE layer) lora fine-tuning doesn't work

Status: Open.

#3080 In huggingface/peft;

· dskhudia opened on Mar 4, 2026

TP support for Finetuning using LoRA and other PEFT techniques

Status: Open.

#3044 In huggingface/peft;

· quic-meetkuma opened on Feb 16, 2026

LoRA + activation checkpointing breaks model compilation with torch.compile when dropout != 0

Status: Open.

#3020 In huggingface/peft;

· raliouat opened on Feb 3, 2026

Combining adapters linearly with negative weights is broken

Status: Open.

#3004 In huggingface/peft;

· orr-hirundo opened on Jan 21, 2026

Add EWoRA to PEFT

Status: Open.

#2970 In huggingface/peft;

· HarshKohli opened on Dec 31, 2025

Return base model state_dict with original keys

wip

Status: Open.

#2945 In huggingface/peft;

· dvmazur opened on Dec 9, 2025

enusre_weight_tying = True does not respect the model config tie_word_embeddings flag

Status: Open.

#2944 In huggingface/peft;

· romitjain opened on Dec 9, 2025

Deal with weight tying consistently

contributions-welcome

Status: Open.

#2864 In huggingface/peft;

· BenjaminBossan opened on Oct 23, 2025