baaivision/EVA
EVA provides a series of vision foundation models including ViT variants, EVA-CLIP, and masked visual representation learning approaches from BAAI.
Not currently ranked — collecting fresh signals.
star history
The EVA repository hosts multiple visual representation models from the Beijing Academy of Artificial Intelligence. It includes EVA-01 (masked visual representation learning at scale), EVA-02 (improved vision transformer architecture), and EVA-CLIP variants (contrastive language-image pretraining) scaling up to 18 billion parameters. The models are available through Hugging Face and timm libraries.
Frequently asked
- What is baaivision/EVA?
- EVA provides a series of vision foundation models including ViT variants, EVA-CLIP, and masked visual representation learning approaches from BAAI.
- Is EVA open source?
- Yes — baaivision/EVA is open source, released under the MIT license.
- What language is EVA written in?
- baaivision/EVA is primarily written in Python.
- How popular is EVA?
- baaivision/EVA has 2.7k stars on GitHub.
- Where can I find EVA?
- baaivision/EVA is on GitHub at https://github.com/baaivision/EVA.