QVAC: Local AI SDK for Peer-to-Peer Applications

QVAC is an open-source, cross-platform ecosystem designed for building local-first, peer-to-peer AI applications and systems. It enables users to run various AI tasks such as LLMs, speech-to-text, and translation locally across multiple operating systems including Linux, macOS, Windows, Android, and iOS.

Key features:

• Local-first - Load AI models and perform inference on your own machine without relying on third-party APIs or cloud services.
• P2P - Build decentralized internet systems similar to BitTorrent or blockchain networks specifically for AI.
• Cross-platform - Consistent developer experience across different hardware and software environments.
• OpenAI-compatible API - Seamlessly integrate with the broader AI ecosystem.
• Open source - Completely free to use and modify, encouraging community contributions.

Developers can utilize the JavaScript SDK as the main entry point, which is type-safe and exposes all QVAC capabilities through a unified interface. The SDK runs on Node.js and provides tools for managing AI models and inference tasks.