TEXT TO SPEECH API
Natural speech from any text.
Turn text into speech with Qwen3 TTS, Kokoro, Orpheus, Chatterbox or Zonos through an OpenAI-compatible endpoint, billed per 1K characters.
Hand this to your agent
Add a Runcrate model to this project: run curl -fsSL"https://runcrate.ai/agents.md?intent=build"and follow it.
Your agent asks for an API key. Create a key
Quick start
Hand it to your agent. Or read the request it sends.
Your agent writes and runs it. The same key works for every model and every GPU.
Available models
Models you can call today. Billed prices, as charged.
| Model | Maker | Price | Detail |
|---|---|---|---|
| Qwen/Qwen3-TTS | Alibaba | $0.021 per 1K characters | Multilingual, natural prosody |
| hexgrad/Kokoro-82M | Hexgrad | $0.00104 per 1K characters | Fast, expressive voice synthesis |
| ResembleAI/chatterbox-multilingual | Resemble AI | $0.00105 per 1K characters | Conversational TTS |
Why Runcrate
Built for production. One key, one balance.
Multiple voice models
Choose from six different TTS engines. Each has different strengths: natural prosody, emotional range, speed, or multilingual coverage.
Per-character billing
Each model lists a price per 1K characters. You pay for the text you send, with no subscription or seat licence.
OpenAI-compatible
Uses the standard /v1/audio/speech endpoint. If your app works with OpenAI TTS, it works here with a base URL change.
Voice customization
Select from preset voices or adjust parameters for speed, pitch, and emotional tone depending on the model.
FAQ