Skip to content

Audio Input

Use the input_audio content-part to send audio to audio-capable models.

Terminal window
curl https://llmtr.com/v1/chat/completions \
-H "Authorization: Bearer sk_your_key" \
-H "Content-Type: application/json" \
-d '{
"model": "google/gemini-2.5-flash",
"messages": [
{
"role": "user",
"content": [
{ "type": "text", "text": "Transcribe and briefly summarize." },
{
"type": "input_audio",
"input_audio": {
"data": "BASE64_WAV_DATA",
"format": "wav"
}
}
]
}
]
}'
  • wav
  • mp3
  • flac
  • ogg
  • webm
  • Inline base64: < 1 MB recommended.
  • Longer audio: upload via Files API and reference file_id.
{
"type": "input_file",
"input_file": {
"file_id": "file_xxx",
"mime_type": "audio/wav"
}
}

Audio input is mainly available on Google Gemini family. Check:

Terminal window
curl "https://llmtr.com/api/models?modality=audio" \
-H "Authorization: Bearer sk_your_key"