🚀Pocket LLM v1.5.0 is out: offline Android LLM chat with voice, image input, OCR, and camera capture

Pocket LLM v1.5.0 is a major upgrade for the open-source Android app that runs large language models entirely offline. The new release introduces voice input, letting users speak queries instead of typing, and image input with OCR (optical character recognition) for extracting text from photos. It also supports Gemma vision and FastVLM models, enabling visual question answering and real-time camera capture with features like retake, crop, and photo review. A side panel now organizes previous chats for easy access, and users can delete downloaded models to reclaim storage space.

Customization gets a boost with editable model instructions, including presets and custom prompts for tailored responses. The app also adds light/dark mode, accent colors, and font-size controls for a personalized interface. Assistant responses can be copied with a single tap. All processing happens locally on the Android device, ensuring complete privacy and offline functionality. This update positions Pocket LLM as a robust, private AI companion for professionals who need on-the-go assistance without cloud dependencies.