tetherto/qvac

Open-source local AI SDK - run AI on-device with no cloud, no API keys. Supports GGUF, RAG, image, music, and video generation, speech-to-text, P2P inference, and more. Cross-platform: Linux, macOS, Windows, Android, iOS.

What it solves

QVAC provides a unified way to run a wide variety of AI workloads locally on commodity hardware across multiple platforms (Linux, macOS, Windows, Android, and iOS). It eliminates the need for cloud-based AI APIs, ensuring data privacy, offline availability, and full user control over the AI system.

How it works

QVAC offers two primary integration paths: an SDK for JavaScript/TypeScript and Python to build local-first AI applications, and an OpenAI-compatible HTTP server that allows existing AI tools to use QVAC as a local model provider. It utilizes optimized open models and features a peer-to-peer (P2P) system for downloading models directly between users, bypassing centralized hosting.

Who it’s for

Developers building mobile and desktop applications that require AI capabilities without cloud dependencies, and users who want to serve local models to their favorite AI tools.

Highlights

  • Cross-platform support: Works across desktop and mobile OSs using a single codebase.
  • Comprehensive AI capabilities: Supports text generation, RAG, fine-tuning (LoRA), multimodal inference, image/video/music generation, transcription, and OCR.
  • P2P Model Distribution: Uses a distributed model registry to fetch models directly from peers.
  • OpenAI-compatible API: Allows seamless integration with tools like OpenCode and OpenClaw via its HTTP server.

Related

  • Project
    SearchSavior/OpenArc

    An OpenVINO-powered inference engine for Intel devices that serves LLMs, VLMs, and audio models via OpenAI-compatible endpoints.

  • Project
    orailnoor/cross-platform-llm-client

    A cross-platform AI chat client that unifies local on-device LLM inference via llama.cpp with cloud API access for OpenAI, Anthropic, and Google Gemini.

  • Project
    AtomicBot-ai/Atomic-Chat

    A local AI application and inference engine that lets users run open-weight LLMs privately on their own devices while providing an OpenAI-compatible API for other agents.

  • Project
    duixcom/Duix-Mobile

    An open-source SDK for deploying real-time interactive AI avatars on mobile and embedded devices with low latency and on-device execution.

  • Project
    hyperspaceai/hyperspace-node

    A decentralized P2P AI inference network that allows users to contribute GPU, CPU, and storage resources to run AI models and earn rewards.