Skip to main content
Google’s production Lyria 2 model via Vertex AI. One ~30s, 48 kHz stereo WAV clip per request. Supports negative prompts and seed for reproducibility.

Request

Required: prompt_tags
Optional: duration, energy, negative_prompt, sample_count, seed, tempo
Reference inputs take asset ids from POST /v1/uploads or an earlier output’s asset_id. Prices and allowed values come from the live catalog; GET /v1/models/lyria-002 is always current.