CLIP
Collection
Multi-modal models that can be used for Smart Search in Immich. Models are sorted by size in descending order.
•
35 items
•
Updated
•
10
This repo contains ONNX exports for the corresponding ResNet-101-based CLIP model by OpenAI. See the CLIP repo for more info. Visual and textual encoders are separated into separate models for the purpose of generating image and text embeddings.
This repo is specifically intended for use with Immich, a self-hosted photo library.