Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.nano-gpt.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Use GET /api/v1/audio-models to discover the currently available audio models. The response includes both text-to-speech (TTS) and speech-to-text (STT) models. This endpoint is cacheable. Refresh it periodically and do not hardcode audio model capabilities in your client.

​
Endpoint

GET https://nano-gpt.com/api/v1/audio-models

​
Authentication

Authentication is optional.
  • Authorization: Bearer YOUR_API_KEY
  • x-api-key: YOUR_API_KEY

​
Query Parameters

ParameterTypeDefaultDescription
detailedbooleantrueInclude names, descriptions, pricing, capabilities, and supported parameters.
typestringallFilter by audio model type. Allowed values: all, tts, stt.

​
Response

{
  "object": "list",
  "data": [
    {
      "id": "tts-model-id",
      "object": "model",
      "name": "Display name",
      "description": "Model description",
      "architecture": {
        "modality": "audio",
        "input_modalities": ["text"],
        "output_modalities": ["audio"]
      },
      "pricing": {
        "currency": "USD"
      },
      "capabilities": {
        "text_to_speech": true,
        "speech_to_text": false
      },
      "supported_parameters": {}
    }
  ],
  "meta": {
    "count": 1,
    "generated_at": "2026-05-07T12:00:00.000Z"
  }
}
When type is not all, meta may include the active filter.

​
Model Types

TypeDescription
ttsText-to-speech models for POST /api/v1/audio/speech and related TTS endpoints.
sttSpeech-to-text models for transcription endpoints.
allBoth TTS and STT models.

​
Example

curl "https://nano-gpt.com/api/v1/audio-models?type=stt&detailed=true" \
  -H "x-api-key: $NANOGPT_API_KEY"

​
Notes

  • Supported parameters vary by model, including voices, formats, streaming support, file formats, language support, diarization, timestamps, and max input size.
  • Use this endpoint instead of hardcoding audio model capabilities.
  • The response is cacheable, but model availability can change.