⚠️ This update includes the following breaking changes for advanced users who have modified their docker-compose.yml file:
- To disable machine learning now, use IMMICH_MACHINE_LEARNING_ENABLED=false (previously IMMICH_MACHINE_LEARNING_URL=false)
- Sentence-Transformers is no longer used for CLIP models, users who set MACHINE_LEARNING_CLIP_IMAGE_MODEL or MACHINE_LEARNING_CLIP_TEXT_MODEL must migrate to one of the ViT-B models listed here, with the caveat that OpenCLIP models will require running CLIP on all images since the embeddings are incompatible.