Edge-Veda: On-Device AI SDK for Flutter with LLM Inference, Vision, and Speech Processing
By
rish2497
Slow-proofed and worth the wait. Worth its weight in flour.
Summary
Edge-Veda is an on-device AI SDK for Flutter that enables local AI inference without cloud dependencies. It supports multiple AI capabilities including LLM inference, vision processing, speech-to-text, text-to-speech, image generation, embeddings, RAG (Retrieval-Augmented Generation), and function calling. The SDK leverages Metal GPU acceleration on iOS/macOS devices and is designed to run sustainably on real phones under real constraints. Currently iOS-only with Android support planned, it offers a managed runtime with approximately 22,700 lines of code, 40 C API functions, and 32 Dart SDK files.
Key quotes
· 5 pulledA managed on-device AI runtime for Flutter — text, vision, speech-to-text, text-to-speech, image generation, and RAG running sustainably on real phones under real constraints.
Private by default.
iOS only (iPhone, Metal GPU). Android support is on the roadmap.
~22,700 LOC | 40 C API functions | 32 Dart SDK files | 0 cloud dependencies
Asking questions about a medical report — RAG retrieval + LLM generation, entirely on-device
You might also wanna read
Google AI Edge Gallery: On-Device Generative AI App for Local Model Testing
Google has launched AI Edge Gallery, an experimental Android app that allows users to run generative AI models locally on their devices with
Product Hunt·8mo ago
NexaSDK for Mobile: On-Device Multimodal AI Deployment for iOS and Android
NexaSDK for Mobile enables developers to deploy multimodal AI models fully on-device for iOS and Android apps, leveraging Apple Neural Engin
Product Hunt·5mo ago
Google Launches AI Edge Gallery for On-Device Generative AI Testing
Google has launched AI Edge Gallery, an experimental Android app that allows users to run generative AI models locally on their devices with
Product Hunt·8mo ago
LFM2: Liquid AI's Hybrid Edge AI Models for On-Device Deployment
LFM2 by Liquid AI is a new generation of hybrid models designed for on-device edge AI applications. The 1.5B parameter model is optimized fo
Product Hunt·4mo ago
Reka Vision Launches Reka Edge: Efficient 7B Vision Language Model for Physical AI
Reka Vision has launched Reka Edge, a highly efficient 7B Vision Language Model designed for Physical AI applications. The model features a
Product Hunt·1mo ago
Lekh AI: On-Device Private AI Assistant for iPhone, iPad, and Mac
Lekh AI is a native iOS and macOS app that runs AI models entirely on-device without cloud connectivity, offering private, offline AI capabi
Product Hunt·4mo ago