Google DeepMind text-to-music. Production-ready.

Google Lyria 3 Pro APIProduction access. 30-second generation.

Google Lyria 3 Pro is Google DeepMind's production text-to-music model. MusicAPI provides direct developer access through a single REST endpoint. Generate, extend, cover, and replace music with studio-grade output, ~30 second turnaround, and flat 12-credit pricing per task.

What is Google Lyria 3 Pro?

Google Lyria 3 Pro is the production release of Google DeepMind's text-to-music model family. It accepts a natural-language prompt and optional lyrics, and returns a studio-quality audio clip with synthesized vocals, instrumentation, or both. Lyria 3 Pro succeeds the earlier Lyria 3 preview and is designed for high musical coherence across genre, tempo, and arrangement.

MusicAPI exposes Lyria 3 Pro through its Producer API namespace at /api/v1/producer/*. Existing customers from the previous Producer model platform continue to work without code changes: legacy mv values (FUZZ-2.0, Lyria 3 preview, and so on) are silently aliased to Lyria 3 Pro.

Lyria 3 Pro at a glance

~30 second generation

Median task latency around 30 seconds. Fast enough for interactive product flows.

Studio-grade output

Quality comparable to Suno v5 across clarity, dynamics, and musical structure.

1 to 240 second clips

Bound clip length with the length field. Combine extend_music for longer tracks.

12 credits per task

Flat pricing. No surcharge by clip length, instrumental vs vocal, or prompt complexity.

Call Google Lyria 3 Pro in one HTTP request

Authenticate with a single Authorization: Bearer header, POST a JSON body describing the sound you want, and receive a task ID immediately. Poll for the finished clip or register an HTTPS webhook for completion notification.

# Generate music with Google Lyria 3 Pro via MusicAPI Producer API
curl -X POST https://api.musicapi.ai/api/v1/producer/create \
  -H "Authorization: Bearer $MUSICAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "task_type": "create_music",
    "sound": "cinematic orchestral trailer with powerful choirs and dramatic strings",
    "title": "Trailer Cue",
    "length": 60
  }'

# Response (immediate):
# { "message": "success", "task_id": "29906f02-2366-4218-b25a-41458e1d60e0" }

# Poll for the finished Lyria 3 Pro clip (about 30 seconds later):
curl -H "Authorization: Bearer $MUSICAPI_KEY" \
  https://api.musicapi.ai/api/v1/producer/task/29906f02-2366-4218-b25a-41458e1d60e0

Four operations on one Lyria 3 Pro endpoint

Switch operations by changing the task_type field. Same authentication, same response shape, same 12-credit cost.

create_music

Generate a new track from a sound prompt and optional lyrics. The primary entry point.

extend_music

Continue an existing clip from a specific timestamp. Pass clip_id, starts_at, and the new musical direction in instruction.

replace_music

Swap a specific time window with new content. Targets a chorus, a vocal segment, or any precise starts_at/ends_at region.

cover_music

Reinterpret a clip in a new style. Adjustable strength parameter (0 to 1) controls how far the cover diverges from the source. Re-enabled May 2026.

Google Lyria 3 Pro vs Suno v5

Both models are available through one MusicAPI key. Use the comparison below to decide which fits the request you are about to send.

DimensionGoogle Lyria 3 ProSuno v5
Typical generation time~30 seconds~60 to 120 seconds
Quality positioningStudio-grade, strong dynamicsStudio-grade, distinctive vocal style
Clip length range1 to 240 seconds per taskUp to 8 minutes via extend
Operationscreate, extend, replace, covercreate, extend, cover, persona, remaster, MIDI, stems
Pricing on MusicAPI12 credits per task10 credits per task (sonic-v5)
Endpoint namespace/api/v1/producer/*/api/v1/sonic/*

What developers ship with Lyria 3 Pro

Game soundtracks

Dynamic in-game music that responds to player state. Generate scene-specific tracks at runtime instead of shipping fixed audio assets.

Video editing tools

Royalty-free background music produced from a one-line description. Replaces stock-music search for short-form video creators.

Music apps and SaaS

User-facing AI music products that generate, edit, and remix tracks. The Producer cover operation is well suited to remix flows.

Podcasts and audiobooks

Custom intro music, transition stings, and chapter beds. Lyria 3 Pro handles instrumental prompts cleanly.

Meditation and wellness

Ambient soundscapes tuned to session length. The 1 to 240 second per-task range lines up with most meditation segments.

Ad and brand audio

Bespoke jingles and ad music produced in seconds. Commercial use is included with every MusicAPI generation.

Google Lyria 3 Pro API: Frequently asked questions

What is Google Lyria 3 Pro?

Google Lyria 3 Pro is Google DeepMind's production text-to-music model. It generates studio-grade vocals, instrumentals, and full tracks from a natural-language prompt, with track durations from 1 to 240 seconds. Lyria 3 Pro succeeds the earlier Lyria 3 preview and is designed for high musical coherence, clarity, and dynamics.

How do I get an API key for Google Lyria 3 Pro?

Sign up at musicapi.ai, generate a MusicAPI key from the dashboard, and call POST /api/v1/producer/create with task_type 'create_music' and your sound or lyrics prompt. The same key works for every Producer operation (create, extend, replace, cover) and is included in the standard MusicAPI pricing.

How long does a Google Lyria 3 Pro generation take?

Around 30 seconds per task in the typical case. Producer API submits the job, returns a task_id immediately, and your application either polls GET /api/v1/producer/task/{task_id} or receives an HTTPS webhook when the clip is ready.

How much does the Google Lyria 3 Pro API cost?

12 credits per generation task on MusicAPI. Pricing is fixed per task and does not scale with clip length, prompt complexity, or whether the result is instrumental or vocal. Upload and download tasks cost 2 credits each. Failed upstream tasks are auto-refunded.

Google Lyria 3 Pro vs Suno v5: which should I use?

Use Google Lyria 3 Pro when you need fast turnaround (~30 seconds) and studio-grade audio with strong dynamics. Use Suno v5 when you need its specific lyric phrasing and stylistic range. Both are available through MusicAPI under one account and one billing surface, so you can A/B them without switching providers.

Which operations does Google Lyria 3 Pro support?

Lyria 3 Pro supports four core operations through Producer API: create_music (new generations), extend_music (continue from a timestamp), replace_music (swap a time window), and cover_music (reinterpret an existing clip in a new style with an adjustable strength parameter). Each uses the same POST /api/v1/producer/create endpoint.

Can I use Google Lyria 3 Pro output commercially?

Yes. All music generated through MusicAPI's Producer API, including Lyria 3 Pro output, ships with commercial rights. You own the generated audio and may use it in games, videos, apps, podcasts, ads, and other commercial productions.

Is there a free trial for the Google Lyria 3 Pro API?

Yes. New MusicAPI accounts receive 75 free credits on signup, which is enough for six full Lyria 3 Pro generation tasks (12 credits each). You can also try the model anonymously, with no account, in the public playground at musicapi.ai/playground.

Start building with Google Lyria 3 Pro

75 free credits on signup. No credit card required to try the API in the playground.