How to use Peft: what you can do with it

3 use cases

• Fine-tune a small LLM with LoRA on one GPU

  Wrap a base model with a LoRA config so you train only a tiny fraction of its parameters.

  • REPO huggingface/peft
  • Claude Code
  • Codex
  • Cursor
  • Software and DevOps
  • Intermediate
  View prompt

  Using huggingface/peft, fine-tune Qwen/Qwen2.5-3B-Instruct with LoRA (r=16, lora_alpha=32) on <dataset> using the Transformers Trainer. Print the trainable parameter count first, save the adapter, and report training loss.

  Share Email me this

• Serve several task adapters from one base model

  Save one small adapter per task and load or switch between them on a single base model.

  • REPO huggingface/peft
  • Claude Code
  • Codex
  • Cursor
  • Software and DevOps
  • Intermediate
  View prompt

  Using huggingface/peft, load base model <model> once and attach adapters <adapter A> and <adapter B> with load_adapter. Write a function that switches adapters with set_adapter per request and test each on <prompt>.

  Share Email me this

• Train a custom Stable Diffusion style with LoRA

  Use PEFT LoRA with Diffusers to teach an image model a subject or style on a modest GPU.

  • REPO huggingface/peft
  • Claude Code
  • Codex
  • Cursor
  • Software and DevOps
  • Advanced
  View prompt

  Using huggingface/peft, run examples/lora_dreambooth/train_dreambooth.py on CompVis/stable-diffusion-v1-4 with my images in <folder>, using LoRA and gradient checkpointing. Report peak GPU memory and save the adapter.

  Share Email me this

Quick start with Peft

1. Check you have what it needs: Python with PyTorch and Transformers; A GPU with enough memory for the base model (the README shows a 3B model on far less memory with LoRA); A dataset for your task.
2. Open huggingface/peft on GitHub and read the README and license.
3. Install with pip install peft
4. Load a base model such as Qwen/Qwen2.5-3B-Instruct with Transformers
5. Create a LoraConfig with r=16 and lora_alpha=32 for a causal LM and wrap the model with get_peft_model
6. Call model.print_trainable_parameters() to confirm only about 0.12% of parameters are trainable

These steps come from our top use case, Fine-tune a small LLM with LoRA on one GPU, which also has a copy-paste agent prompt.

Using the Deduped MCP, get the setup steps for huggingface/peft

claude mcp add --transport http deduped https://dedupedai.com/mcp

Best AI repos for Software and DevOps

About this repo

• Who it is forML engineers and hobbyists who want to adapt open models to their own data on a single GPU.
• Why it is hot22k stars, talked about in 2 AI sources we track
• LanguagePython
• Sourcegithub.com/huggingface/peft · Back to Hot repos

Requirements

• Python with PyTorch and Transformers
• A GPU with enough memory for the base model (the README shows a 3B model on far less memory with LoRA)
• A dataset for your task

Related repos

• Trended earlier REPO huggingface/trl TRL TRL is Hugging Face's library for post-training foundation models. It provides trainers for Supervised Fine-Tuning (SFT), GRPO, DPO, KTO and reward modeling on top of Transformers, scales from one GPU to multi-node clusters with Accelerate, integrates PEFT for LoRA and QLoRA, and ships a CLI that fine-tunes without writing code.
• Trended earlier REPO whitecircle/halo Halo Halo is an open-source framework from White Circle for training large language and multimodal models with pre-training, SFT, preference optimization, and multi-turn RL on Hugging Face models.
• Trended earlier REPO EleutherAI/lm-evaluation-harness LM Evaluation Harness The LM Evaluation Harness is EleutherAI's framework for few-shot evaluation of language models. One lm_eval command scores a model on named tasks, with backends for Hugging Face transformers, vLLM, SGLang, llama.cpp servers, ONNX Runtime, NeMo, Megatron and any OpenAI compatible API. It supports multi GPU evaluation, result saving, response logging and pushing results to the Hugging Face Hub.