QVAC Provider

QVAC is an open-source, cross-platform runtime for local-first, peer-to-peer AI — LLMs, embeddings, transcription, translation, speech, and image generation, all running on the user's own hardware.

@qvac/ai-sdk-provider is a branded wrapper around @ai-sdk/openai-compatible that points at a local qvac serve openai HTTP server and re-exports QVAC's typed model metadata.

This provider talks to a local OpenAI-compatible server. Install @qvac/cli and start it with qvac serve openai before using the provider.

Setup

The QVAC provider is available in the @qvac/ai-sdk-provider module. You can install it with:

pnpm add @qvac/ai-sdk-provider ai @ai-sdk/openai-compatible

ai and @ai-sdk/openai-compatible are peer dependencies — install them alongside.

Provider Instance

Import createQvac from @qvac/ai-sdk-provider and create a provider instance pointing at your running qvac serve openai:

import { createQvac } from '@qvac/ai-sdk-provider';


const qvac = createQvac({
  baseURL: 'http://127.0.0.1:11434/v1', // match your `qvac serve` port
  apiKey: 'qvac', // anything non-empty; serve does not validate
});

Language Models

You can create models by passing a qvac serve alias to the provider instance:

import { createQvac } from '@qvac/ai-sdk-provider';
import { generateText } from 'ai';


const qvac = createQvac({
  baseURL: 'http://127.0.0.1:11434/v1',
  apiKey: 'qvac',
});


const { text } = await generateText({
  model: qvac('qwen3.5-0.8b'),
  prompt: 'Write a haiku about local-first AI.',
});


console.log(text);

Additional Resources