GerbilGerbil
Browser Runtime

Playground

Run Gerbil directly in your browser. No installation required.

Requires WebGPU support (Chrome 113+, Edge 113+)

Loading playground...

How it works

  • 01Models are downloaded to your browser's cache on first use (one-time download)
  • 02Inference runs locally using WebGPU for GPU acceleration
  • 03Your prompts never leave your device — completely private

Text

  • smollm2-135m — 100MB
  • smollm2-360m — 250MB
  • qwen3-0.6b — 400MB

Vision

  • ministral-3b — 2.5GB
  • Image understanding
  • OCR & analysis

Speech

  • Kokoro-82M — 330MB
  • Supertonic-66M — 130MB
  • 28+ voices (US/UK)

Transcribe

  • whisper-tiny — 39MB
  • whisper-base — 74MB
  • 44x realtime

Tips

  • —Use Cmd+Enter to generate/speak
  • —Enable thinking mode for math and reasoning
  • —First use downloads models (cached after)
  • —All processing happens on-device

// Want to use Gerbil in Node.js? Check out the docs