Edge-Veda
A Dart SDK that runs AI models directly on iPhones and Macs, covering chat, vision, speech-to-text, speech synthesis, image generation and retrieval, entirely offline.
Edge-Veda is a managed on-device AI runtime for Flutter apps on Apple hardware. It runs large language model inference, vision, speech-to-text, text-to-speech, image generation and retrieval-augmented generation entirely on an iPhone or Mac's Metal GPU, with no cloud dependency and no data leaving the device.
The project's stated purpose is reliability rather than just running a demo: it manages thermal throttling and memory pressure so long sessions stay stable, and gives developers visibility into runtime behavior instead of a black box. It ships as a C API of about 40 functions wrapped in a Dart SDK of 32 files, roughly 22,700 lines of code.
It currently supports iOS only, with Android support planned. A companion tool automates project setup, model selection and device deployment.