Suno API documentation — generate music from a prompt (V5 / V5.5), with pricing

Suno is a best-in-class AI music model — it generates full songs with vocals, lyrics and instrumentation from a text prompt. The Suno API (Suno V5 and V5.5 on Kunavo) does this programmatically via POST /v1/audio/music, returning hosted audio tracks on one bearer token. This is the practical documentation: the request shape, pricing, and the async pattern for production.

Call the Suno API

Generate a key at /app/keys and POST a prompt. One request returns about two track variations; generation takes tens of seconds to a couple of minutes:

import requests

# Music generation is a POST to /v1/audio/music. One request returns
# ~2 track variations; generation takes tens of seconds to a couple of minutes.
resp = requests.post(
    "https://api.kunavo.com/v1/audio/music",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={
        "model": "suno-v5",                 # or suno-v5-5
        "prompt": "uplifting lo-fi hip hop with mellow piano and vinyl crackle",
        "instrumental": False,              # set True to skip vocals
    },
    timeout=300,
)
for track in resp.json()["data"]:
    print(track["url"])

Full parameters and the async task lifecycle are in the music API reference.

Pricing

Suno is billed per generation request (not per track), so the ~2 variations you get back cost one charge. Live rates are on each model page.

Pay-as-you-go from a $5 minimum top-up, balance never expires, failed generations never billed. Estimate a broader workload with the cost calculator.

Custom mode: your own lyrics

Instead of a plain prompt, supply lyrics, a style descriptor and a title for full control, or set instrumental: true to skip vocals entirely:

# Custom mode: supply your own lyrics + a style, instead of a plain prompt.
curl https://api.kunavo.com/v1/audio/music \
  -H "Authorization: Bearer $KUNAVO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "suno-v5-5",
    "lyrics": "[Verse]\nCity lights blur into the rain...",
    "style": "synthwave, 110 bpm, female vocal",
    "title": "Neon Rain"
  }'

Async in production

Music generation isn't instant, so at scale submit a task and poll for the result rather than holding the connection — the music docs cover the task lifecycle, polling and webhook delivery. Result URLs are permanent once complete.

Suno API vs Udio API

Udio is the other leading music-generation family, and "Udio API" is a common parallel search. The practical difference today: Udio's API access runs through experimental third-party paths without a stable public price, while Suno V5 / V5.5 on Kunavo is a supported endpoint with a flat, published rate ($0.09 per generation request). If you need dependable music generation in production — the Suno V5 API here is the reliable route; we'll evaluate Udio once a stable upstream exists.

Beyond music

The same key reaches text (Claude, Gemini, GPT), image (Nano Banana, GPT-Image-2) and video (Veo 3, Seedance 2, Wan 2.7) — so you can chain a full pipeline: script with an LLM, generate visuals, add a Suno soundtrack. See the multimodal AI guide for image → video → music workflows.

FAQ

What is the Suno API?

A programmatic way to generate full songs — vocals, lyrics and instrumentation — via POST /v1/audio/music. Kunavo serves Suno V5 and V5.5.

How much does the Suno API cost?

$0.09 per generation request on Kunavo (about 20% under Suno's list), returning ~2 track variations for one charge.

How do I get a Suno API key?

Sign up, create a key at /app/keys, set base_url=https://api.kunavo.com/v1. The quickstart has the first call.

Can I make instrumental music or use my own lyrics?

Yes — instrumental: true for no vocals, or custom mode with your own lyrics and style. Both hit /v1/audio/music.