Documentation

Install Wfloat, then choose a model. Model pages provide the shortest working example; these pages describe the SDK API.

• Model management: downloads, progress, storage, and cleanup.
• Language model: messages, generation, tools, and structured output.
• Text to speech: audio generation, dialogue, and playback.
• Speech to text: recordings and live transcription.
• Voice activity detection: speech ranges and live detection.

Choose Web, React Native, or Python at the top of a reference page. Your selection carries across the docs.