Suno AI API Specifications
High-throughput AI music generation infrastructure. Explore endpoints, supported models, asynchronous task lifecycles, and developer best practices.
What is the Suno AI API?
The Suno AI API is a cloud-based interface enabling automated composition, lyric generation, vocal arrangement, and music synthesis. Through SongMesh's RESTful architecture, developers can programmatically trigger song creations that return two complete variations per generation prompt, complete with cover art, streamable audio URLs, and timeline metadata.
Supported AI Models
| Model ID | Best For | Generation Time | Default |
|---|---|---|---|
| V6 | Standard production quality, balanced vocal clarity, and complex arrangement. | 35-50s | Yes |
| V6_WILD | Experimental styles, unconventional song structures, and innovative instrumentation. | 40-55s | No |
| V6_MINI | Rapid prototyping, interactive games, and fast preview generations. | 20-30s | No |
Asynchronous Task Lifecycle
Submit Task
Client sends POST /api/v1/createTask with prompt, style tags, and optional callBackUrl. Receives taskId immediately.
GPU Inference
Distributed cloud inference clusters synthesize stereo audio, lyrics alignment, and cover visuals in parallel.
Webhook Dispatch
Our gateway dispatches an HTTP POST callback to your endpoint with signed HMAC headers containing finished audio URLs.
Python SDK / Requests Example
requests libraryimport requests
url = "https://sunoapi.top/api/v1/createTask"
headers = {
"Authorization": "Bearer sk_live_your_secret_key",
"Content-Type": "application/json"
}
payload = {
"taskType": "generate_music",
"prompt": "Lo-fi hip hop beat with calm vinyl crackle and soothing rhodes piano",
"tags": "lofi, chillhop, study, calm",
"customMode": True,
"instrumental": True,
"model": "V6"
}
response = requests.post(url, headers=headers, json=payload)
data = response.json()
print("Task created successfully, Task ID:", data.get("data", {}).get("taskId"))Frequently Asked Questions
What AI models are supported by the Suno AI API?
SongMesh supports the latest V6 audio synthesis models, including V6 (balanced default for broadcast fidelity), V6_WILD (maximum creative variation and experimental rhythms), and V6_MINI (optimized for low-latency preview generations).
How does asynchronous generation work in the Suno AI API?
Because generating studio-quality dual-track audio takes 30-60 seconds on cloud GPU clusters, the API uses an asynchronous job architecture. Your POST /api/v1/createTask request immediately returns a taskId. When the track is ready, our system delivers the final audio URLs to your callBackUrl via HTTP POST, or you can query /api/v1/queryTasks.
What audio format and sample rate are produced?
Generated songs are rendered in high-bitrate stereo MP3 (320kbps, 44.1kHz). You can also use our WAV conversion endpoint to obtain broadcast-standard uncompressed 44.1kHz / 16-bit PCM WAV master files.
Can I extract instrumental and vocal stems separately?
Yes. The Suno AI API provides dedicated stem separation endpoints (separate_vocals) that extract isolated vocal stems and instrumental backings with high phase coherence.
Related Documentation
Start Generating with the Suno AI API
No monthly commitments. Simple, deterministic pay-as-you-go credit billing with permanent credit validity.
SongMesh (sunoapi.top) is an independent third-party developer integration platform. We are not affiliated with, endorsed by, or sponsored by Suno, Inc. "Suno" is a registered trademark of Suno, Inc.