This release .....

New Features 🌟

Image search can work with any image locally or form URL
Support for Apple Silicon GPU
Parallel Image indexing and inference for CLIP and OCR
Separation of indexing and searching scripts for memory optimization
Voidtool Everything is no longer required, it's optional now
text embeddings now support hugging face transformers, Llama.cpp, Ollama and any OpenAI Compatible API
top k and threshold selection for all functionalities
Settings UI
Powertoys run Plugin for Windows
Raycast Extension for Mac

Bug Fixes 🐞

If new image is saved with the same name as previous image e.g., untitlied will be reindexed if deep_scan is activated from config
Fixing an issue when ollama isn't running