News › Gemini text-to-speech on the OpenAI audio endpoint

Gemini text-to-speech on the OpenAI audio endpoint

Model launch · 2026-10-15

Google's Gemini text-to-speech models are now available on OneFast:

  • gemini-3.8-flash-tts
  • gemini-3.8-flash-lite-tts

Call them through the OpenAI-compatible POST /v1/audio/speech endpoint — the same request format as OpenAI TTS, so existing OpenAI SDK code works by changing model.

cURL

curl https://api.onefast.ai/v1/audio/speech \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gemini-3.8-flash-tts", "input": "Hello from OneFast.", "voice": "Kore", "response_format": "wav"}' \
  --output hello.wav

Notes:

  • Audio is returned as WAV (or raw PCM). Requests for other formats return WAV.
  • Billing is per token; Gemini audio output is counted at about 25 tokens per second of speech.

See Audio & embeddings for the full list of speech models and live prices in the model catalog.