Prices are per audio-hour of input. Speed is the median over real ModelRelay requests, in seconds of audio transcribed per second. Accuracy is not measured yet, so models are listed rather than ranked.
| Model | Price | Speed × real time, ModelRelay median |
|---|---|---|
| Grok Voice Transcribe 2.0 | $0.10/audio-hour billed per ms of input audio | — |
| Whisper Large v3 Turbo | $0.0024/audio-hour billed per ms of input audio | — |
Use ModelRelay's Responses API directly, or point an existing OpenAI or Anthropic SDK at the compatible endpoints.
135 successful streams measured · 30m window
curl https://api.modelrelay.ai/api/v1/responses \
-H "Authorization: Bearer $MODELRELAY_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"claude-sonnet-5","input":"Hello"}'Change the model. Keep everything else.
Open to everyone. Sign in with GitHub or Google to get started.