QVAC SDK
Use the JS/TS SDK to build local AI applications and systems.
Overview
The JS SDK is cross-platform, type-safe, and exposes all QVAC capabilities through a unified interface.
Key features
- Unified interface: multiple AI tasks, one single npm package to install in your project.
- Cross-platform: consistent behavior across platforms and JS runtime environments — write code once, run it everywhere.
- Type-safe: typed JS API.
Installation
Install and run a quickstart example
Supported platforms, and how to install and run a simple example using the JS/TS SDK.
Functionalities
AI tasks
- Completion: LLM inference via
llama.cpp. - Transcription: speech-to-text (ASR) via
whisper.cpp. - Text embeddings: via
llama.cpp. - Translation: translate text between different languages.
- Text-to-Speech: TTS via ONNX.
- Multimodal: Multimodal via
llama.cpp— i.e., process and understand multiple types of media within the same conversation context. - RAG: out-of-the-box retrieval-augmented generation workflow.
- Delegated inference: perform peer-to-peer inference delegation via Holepunch stack, enabling resource sharing.
Utilities
- Logging: out-of-the-box logging from model to your client.
- Download Lifecycle: pause and resume model downloads.
- Blind relays: establish peer connections through NAT/firewalls by routing traffic via relay nodes.
- Sharded models: download a model that is sharded into multiple parts.
API
API reference
@qvac/sdk npm package exposes a function-centric, typed JS API.
How it works
How it works
Understand what happens under the hood when you use QVAC SDK in your application