Lyria 3.5 - MixRoute API

Google music-generation model for instrumental tracks and songs with vocals.

Model Information

Field	Description
`model`	lyria-3.5
`context`	131,072 input tokens
`endpoint`	POST /v1/models/lyria-3.5:generateContent

Context and output sizes describe the model specification. Effective request limits, availability, and billing depend on the MixRoute route.

Capabilities and Usage

Accepts text and image prompts and returns audio with accompanying text when applicable. It is not a speech transcription or text-to-speech model.
Describe genre, instruments, vocals, and desired duration in the prompt. Duration is not a fixed seconds parameter.
Read candidates[].content.parts[].inlineData.data as Base64 audio and inspect inlineData.mimeType. Default audio is MP3; generated audio is 44.1 kHz stereo.

Request Example

Set MIXROUTE_API_KEY before making the request.

curl --request POST "https://api.mixroute.ai/v1/models/lyria-3.5:generateContent" \
  --header "Authorization: Bearer $MIXROUTE_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "contents": [
    {
      "role": "user",
      "parts": [
        {
          "text": "Create a short instrumental piano track. No vocals."
        }
      ]
    }
  ]
}'

Request Fields

Field	Description
`contents`	Required user content with text and optional image parts.
`generationConfig`	Optional native Gemini generation settings.

Complete request format: gemini-native.