open_clip and clip_playground
The highly popular and downloaded open-source CLIP implementation (A) serves as a foundational library that the demonstrative and less-used notebooks (B) would likely utilize to showcase CLIP's zero-shot capabilities; thus, they are complements.
About open_clip
mlfoundations/open_clip
An open source implementation of CLIP.
Supports diverse Vision Transformer and ConvNet architectures trained on large-scale datasets (LAION-2B, DataComp-1B) with published scaling laws, achieving competitive zero-shot ImageNet accuracy up to 85.4%. Integrates with PyTorch, Hugging Face model hub, and timm for image encoders, enabling efficient embedding computation via the clip-retrieval library. Offers flexible model loading from local checkpoints or HuggingFace, with pre-trained weights optimized for both inference and fine-tuning workflows.
About clip_playground
kevinzakka/clip_playground
An ever-growing playground of notebooks showcasing CLIP's impressive zero-shot capabilities
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work