Software Development & Delivery · Preview

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Best fitUsers who want source-backed project understanding before installing it.

Check whether this project matches your task before installing it.

What it can doprompt, recipe, host_instruction, eval, preflight

Review the portable capability path.

Before continuingVerify in a sandbox

Do not treat a preview pack as a proven local install.

GitHub snapshot21k stars

2.3k forks · 295 contributors

Official first step Read manual preview Source repository

Preview status · 2026-05-16

What is peft?

PEFT (Parameter-Efficient Fine-Tuning) is a Python library developed by Hugging Face that provides efficient methods for fine-tuning pre-trained models while keeping most model parameters ...
Best fit: Users who want source-backed project understanding before installing it.
Capability added to an AI workflow: prompt, recipe, host_instruction, eval, preflight
Evidence base: https://github.com/huggingface/peft, https://github.com/huggingface/peft, https://github.com/huggingface/peft#readme
Preview pages are noindex until English quality, canonical, and citation gates pass.
peft still needs sandbox verification before production use.

Quick decision

Use this section to decide whether the project is worth a deeper read.

Best forUsers who want source-backed project understanding before installing it.

Match the project to your task before installing it.

Capabilityprompt, recipe, host_instruction, eval, preflight

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Repositoryhuggingface/peft

21k stars · 2.3k forks

What it can do

Translate the upstream project into concrete capabilities the user can judge before installing.

Introduction to PEFT

Related topics: Installation Guide, System Architecture, LoRA and LoRA Variants

Sources: [src/peft/tuners/lora/model.py:1-50]()

Installation Guide

Related topics: Introduction to PEFT, Quantization Integration

Sources: [pyproject.toml](https://github.com/huggingface/peft/blob/main/pyproject.toml)

System Architecture

Related topics: Core Components, Introduction to PEFT, Configuration System

Sources: [src/peft/peft_model.py:1-100]()

Core Components

Related topics: System Architecture, Configuration System, Model Loading and Saving

Sources: [src/peft/peft_model.py:1-50]()

LoRA and LoRA Variants

Related topics: Other PEFT Methods, Quantization Integration, Configuration System

Sources: [src/peft/tuners/lora/model.py:1-100]()

Sources: https://github.com/huggingface/peft, Human Manual, Project Pack evidence, and downstream validation signals.

Community Discussion Evidence

Project-level external discussion stays visible on the detail page, not only inside the manual.

Stars21k stars

Forks2.3k forks

Contributors295 contributors

Licenseunknown

Community Discussion Evidence

12 source-linked items

Review these external discussions before using peft with real data or production workflows. They are review inputs, not standalone proof that the project is production-ready.

01
Feature Request: Improve offline support for custom architectures in get
github / github_issue
02
Applying Dora to o_proj of Meta-Llama-3.1-8B results in NaN
github / github_issue
03
Comparison of Different Fine-Tuning Techniques for Conversational AI
github / github_issue
04
[BUG] peft 0.19 target_modules (str) use `set`
github / github_issue
05
v0.19.1
github / github_release
06
v0.19.0
github / github_release
07
0.18.1
github / github_release
08
0.18.0: RoAd, ALoRA, Arrow, WaveFT, DeLoRA, OSF, and more
github / github_release
09
0.17.1
github / github_release
10
0.17.0: SHiRA, MiSS, LoRA for MoE, and more
github / github_release
11
0.16.0: LoRA-FA, RandLoRA, C³A, and much more
github / github_release
12
v0.15.2
github / github_release

How to start

Only source-backed commands are shown here. Verify them in an isolated environment first.

Try the prompt first

Test the workflow without installing the upstream project.

preview

Read the Human Manual

Understand inputs, outputs, limits, and failure modes.

manual

Take context to your AI host

Use the compiled assets in your preferred AI environment.

context

Run sandbox verification

Confirm install commands and rollback before using a primary environment.

verify

pip install peft

Official start command · https://github.com/huggingface/peft#readme · verified: yes

Human Manual

The English page must expose the real manual, not a short placeholder.

8+ sections · Human Manual

peft Manual

PEFT (Parameter-Efficient Fine-Tuning) is a Python library developed by Hugging Face that provides efficient methods for fine-tuning pre-trained models while keeping most model parameters ...

Open the full manual

peft Human Manual
Table of Contents
Introduction to PEFT
Related Pages
Overview
Core Architecture
Design Philosophy
Component Hierarchy

Introduction to PEFT

Related topics: Installation Guide, System Architecture, LoRA and LoRA Variants

Sources: [src/peft/tuners/lora/model.py:1-50]()

Installation Guide

Related topics: Introduction to PEFT, Quantization Integration

Sources: [pyproject.toml](https://github.com/huggingface/peft/blob/main/pyproject.toml)

System Architecture

Related topics: Core Components, Introduction to PEFT, Configuration System

Sources: [src/peft/peft_model.py:1-100]()

Core Components

Related topics: System Architecture, Configuration System, Model Loading and Saving

Sources: [src/peft/peft_model.py:1-50]()

LoRA and LoRA Variants

Related topics: Other PEFT Methods, Quantization Integration, Configuration System

Sources: [src/peft/tuners/lora/model.py:1-100]()

AI Context Pack and portable assets

After deciding to continue, take the project context into your own AI host.

Complete pack plus user-owned assets

These files are planning and verification assets for Claude Code, Codex, Gemini, Cursor, ChatGPT, and other AI hosts.

Download complete pack Read Human Manual

BundleComplete Project Pack AssetAI Context Pack AssetBoundary & Risk Card AssetHuman Manual AssetPitfall Log AssetPrompt Preview AssetQuick Start ManifestPROJECT_PACK.json

Preflight checks

Treat this preview as a planning asset, not proof that your local environment is ready.

The manual is generated from source-linked project files and Doramagic validation signals.
Community evidence warnings stay visible instead of being converted into marketing claims.
The preview remains noindex until English quality and reciprocal indexing gates are explicitly opened.
Use the upstream repository as the final authority for installation commands, license, and version-specific behavior.