4 Pull requests merged by 2 people

• FEAT Allow LoRA to target nn.Parameter

  #2638 merged Jul 15, 2025

• FIX: Error in workflow file to deploy method comparison app

  #2645 merged Jul 14, 2025

• FIX: Prompt learning methods modules_to_save issue

  #2646 merged Jul 14, 2025

• SHiRA Adapters

  #2584 merged Jul 14, 2025

4 Pull requests opened by 4 people

• add a asdict for dataclass model_config

  #2641 opened Jul 11, 2025

• GPT2 compatible version of LLama-Adapters

  #2643 opened Jul 13, 2025

• Add Arrow + GenKnowSub to LoRA

  #2644 opened Jul 13, 2025

• Fix not detecting regex-targeted embedding layer

  #2649 opened Jul 15, 2025

5 Issues closed by 2 people

• GraLoRA support?

  #2558 closed Jul 15, 2025

• Sparse High Rank Adapters (SHiRA) Support

  #2585 closed Jul 15, 2025

• Multi-Lora training

  #2639 closed Jul 14, 2025

• GraniteMoE’s implementation is not compatible with HF’s peft

  #2545 closed Jul 13, 2025

• Call for contribution: KaSA: Knowledge-Aware Singular-Value Adaptation of Large Language Models

  #2516 closed Jul 13, 2025

5 Issues opened by 5 people

• AttributeError: 'DTensor' object has no attribute 'compress_statistics'

  #2650 opened Jul 16, 2025

• Feature Request: Add Adaptive Singular Value Decomposition based Orthogonal Subspace Fine-Tuning

  #2648 opened Jul 15, 2025

• How can I merge the original model weights with LoRA weights?

  #2647 opened Jul 15, 2025

• Prompt_Tuning.ipynb example doesn't seem to train the model

  #2642 opened Jul 11, 2025

• Why peft.utils.other.fsdp_auto_wrap_policy do not warp the module do not require grad?

  #2640 opened Jul 10, 2025

11 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Support for Activated LoRA (Issue https://github.com/huggingface/peft/issues/2523)

  #2609 commented on Jul 14, 2025 • 12 new comments

• [WIP] ENH: Adapter injection based on state_dict

  #2637 commented on Jul 10, 2025 • 3 new comments

• Lora transfer learning

  #2583 commented on Jul 11, 2025 • 0 new comments

• LoRA for experts layers in MoE

  #2527 commented on Jul 11, 2025 • 0 new comments

• How can I fine-tune the linear layers of the LLM part in Qwen2.5_VL 3B?

  #2615 commented on Jul 15, 2025 • 0 new comments

• Make Task Type validation case insensitive

  #2613 commented on Jul 15, 2025 • 0 new comments

• RuntimeError when combining FSDP and disable_adapter

  #1442 commented on Jul 16, 2025 • 0 new comments

• WaveFT method added into tuners

  #2560 commented on Jul 16, 2025 • 0 new comments

• add GLORA

  #2568 commented on Jul 15, 2025 • 0 new comments

• Fix Typos in Test Comments

  #2582 commented on Jul 11, 2025 • 0 new comments

• make method comparison device agnostic, so it can expand to more accelerators like XPU

  #2610 commented on Jul 14, 2025 • 0 new comments