huggingface/peft
peft
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
Builder
HuggingFace
huggingface • ai-lab
Stars
20,879
Using upstream star count
Forks
2,233
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Nov 25, 2022
Project creation date
README Summary
PEFT (Parameter-Efficient Fine-Tuning) is a library that provides state-of-the-art methods for adapting pretrained models to downstream tasks without fine-tuning all model parameters. It supports various techniques like LoRA, AdaLoRA, and other efficient adaptation methods that significantly reduce computational requirements and memory usage. The library integrates seamlessly with Hugging Face Transformers and enables efficient training and inference of adapted models.
AI Dev Skills
Unmapped
Tags
Taxonomy
Deployment Context
Modalities
Skill Areas
Recent Activity
Updated 26 days ago
7 Days
0
30 Days
0
90 Days
0
Quality
production- Quality
- high
- Maturity
- production
Categories
PM Skills
Languages
Timeline
- Project created
- Nov 25, 2022
- Forked
- Mar 22, 2026
- Your last push
- 26 days ago
- Upstream last push
- 11 days ago
- Tracked since
- Mar 18, 2026
Similar Repos
pgvector cosine similarity · $0
Loading…