Search results

Proposal: Benchmarking on different PEFT techniques

Status: Open.

#3363 In huggingface/peft;

· jeet-Abhi123 opened on Jun 27, 2026

Adding support of DEFT, a decompositional efficient fine-tuning framework

Status: Open.

#3349 In huggingface/peft;

· MAXNORM8650 opened on Jun 22, 2026

Only one target_parameters adapter per model

Status: Open.

#3340 In huggingface/peft;

· qgallouedec opened on Jun 18, 2026

LoraConfig(bias="lora_only") trains base_layer.bias but save_pretrained/get_peft_model_state_dict drops it

Status: Open.

#3306 In huggingface/peft;

· hlc1209 opened on Jun 7, 2026

API check: PEFT LoRA merge in an external regression example

Status: Open.

#3305 In huggingface/peft;

· MLPulse opened on Jun 5, 2026

Proposal: add FineGates structured-sparsity PEFT method

Status: Open.

#3292 In huggingface/peft;

· Ofirlin opened on May 30, 2026

Proposal: Content-addressed adapter recipes for cross-platform LoRA distribution

Status: Open.

#3283 In huggingface/peft;

· shiaho777 opened on May 28, 2026

prepare_model_for_kbit_training adds ~1 GB CUDA reserved memory in 500 ms — undocumented cost that breaks memory-constrained training on 8 GB unified-memory devices

Status: Open.

#3265 In huggingface/peft;

· neil-the-nowledgeable opened on May 26, 2026

[BUG] peft 0.19 target_modules (str) use set

Status: Open.

#3229 In huggingface/peft;

· Jintao-Huang opened on May 12, 2026

Proposal: support ShadowPEFT method in PEFT

Status: Open.

#3184 In huggingface/peft;

· SeanLee97 opened on Apr 22, 2026

RFC: Improve code to resolve LoRA variants

wip

Status: Open.

#3182 In huggingface/peft;

· BenjaminBossan opened on Apr 21, 2026

Return base model state_dict with original keys

wip

Status: Open.

#2945 In huggingface/peft;

· dvmazur opened on Dec 9, 2025