Practical, runnable TypeScript recipes for local-first, on-device AI with the QVAC SDK—streaming, RAG, speech, translation, tool calling, and more.
-
Updated
Jul 20, 2026 - TypeScript
Practical, runnable TypeScript recipes for local-first, on-device AI with the QVAC SDK—streaming, RAG, speech, translation, tool calling, and more.
🏛️ Offline multi-agent document council — 3 AI agents debate your documents to a cited answer. The visible disagreement IS the trust mechanism. Built for QVAC Edge AI Hackathon.
Ninety helps away football fans navigate foreign stadiums without relying on cloud services. It uses QVAC for fully on-device OCR and translation, Pears for serverless peer-to-peer sharing of local fan tips, and WDK for self-custodial USDt payments between nearby fans. Built for matchdays where connectivity is unreliable but information and payment
Temporary offline infrastructure for crowds. Phones mesh together (Pears/Hyperswarm) to share hyperlocal tips, translated and ranked by on-device AI (QVAC), with peer-to-peer payments settled via a self-custodial wallet (WDK), no server, no signal required.
Add a description, image, and links to the qvac-sdk topic page so that developers can more easily learn about it.
To associate your repository with the qvac-sdk topic, visit your repo's landing page and select "manage topics."