news
Immich 2.3 Adds Multilingual OCR Models
Quoting: Immich 2.3 Adds Multilingual OCR Models —
The team behind Immich, a popular open-source, self-hosted photo and video management platform, has released version 2.3, delivering a wide range of enhancements across OCR, mobile usability, synchronization, and server management.
The most visible change is the expansion of Immich’s OCR capabilities. Version 2.3 adds new language-specific model variants, bringing support for Greek, Korean, Russian, Belarusian, Ukrainian, Thai, and additional Latin-script languages. Plus, the web viewer now includes an OCR toggle that reveals detected text and allows direct copying.
On mobile, the release adds a new Add/Move action in the asset viewer. The update also improves metadata handling by displaying lens model information in the detail panel, adds a quick date picker for searching, and introduces several refinements to the activity view and deep-link routing.