lucidrains/vit-pytorch
🟢Worth installingStrong adoption and recent activity(Strong adoption and recent activity)
Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
Why this repo matters
- •Canonical PyTorch Vision Transformer implementation serving as the standard reference for attention-based image classification research.
- •High star count indicates sustained community trust and widespread adoption as a foundational computer vision building block.
- •Simplifies SOTA vision model experimentation with a clean, single encoder architecture suitable for rapid prototyping.
Key Metrics
- Stars: 25.5k
- Forks: 3.5k
- Open Issues: 142
- Stars (7d delta): 77
- Stars (30d delta): 77
Scores & Metadata
- Trend Velocity Score: 44.44
- Opportunity Score: 56.53
- Confidence Score: 100.00
- Language: Python
- License: MIT
- Topics:
- First Seen: 4 months ago
Star History
25.4k stars+126 over 23 snapshots6/14/2026 – 7/25/2026