# MusicGen - Core.Today AI API

> Meta's MusicGen via Replicate. Generates music from a text prompt or a reference melody (stereo / melody / large variants), can continue an input clip, and outputs MP3 or WAV. Billed per second of requested duration.

- **Provider**: Meta
- **Model ID**: meta/musicgen
- **Category**: Audio & TTS
- **Credits**: 94 per 8s — 30.5 + 7.9 × seconds (measured), 30 s about 270
- **Speed**: Medium
- **Quality**: Standard

## Features
- Text-to-music and melody-conditioned generation
- Continuation of an input audio clip
- Four model variants (stereo-melody-large, stereo-large, melody-large, large)
- Up to 30 seconds per generation
- Sampling controls (top_k, top_p, temperature, CFG)

## Use Cases
- Short instrumental loops and beds
- Variations on an existing melody
- Extending a clip with continuation

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **prompt**: string - A description of the music you want to generate

### Optional
- **model_version**: string (default: stereo-melody-large) - Model variant. stereo-* produce stereo; melody-* accept input_audio as a melody reference
  Options: stereo-melody-large, stereo-large, melody-large, large
- **duration**: integer (default: 8) - Duration of the generated audio in seconds (1-30). Billed per second
- **input_audio**: string - Audio file that influences the output: continued when continuation is true, otherwise its melody is mimicked (melody models)
- **continuation**: boolean (default: false) - If true, generated music continues from input_audio
- **continuation_start**: integer (default: 0) - Start time (s) of input_audio for continuation
- **continuation_end**: integer - End time (s) of input_audio for continuation. Omit for the end of the clip
- **multi_band_diffusion**: boolean (default: false) - Decode EnCodec tokens with MultiBand Diffusion (non-stereo models only)
- **normalization_strategy**: string (default: loudness) - Strategy for normalizing audio
  Options: loudness, clip, peak, rms
- **top_k**: integer (default: 250) - Reduces sampling to the k most likely tokens
- **top_p**: number (default: 0) - Nucleus sampling probability. 0 uses top_k sampling
- **temperature**: number (default: 1) - Sampling temperature. Higher means more diversity
- **classifier_free_guidance**: integer (default: 3) - Influence of inputs on the output. Higher adheres more to the prompt
- **output_format**: string (default: wav) - Output format for generated audio
  Options: wav, mp3
- **seed**: integer - Seed for the random number generator. Omit or -1 for random

## Examples

### Core.Today theme (instrumental)
The same company-theme prompt given to every instrumental model — compare the results side by side
```json
{
  "model": "meta/musicgen",
  "input": {
    "prompt": "Upbeat K-pop electro-pop company jingle instrumental, 128 BPM, synth brass stabs, punchy drums, bright synth lead, catchy hook, fun and playful",
    "model_version": "stereo-melody-large",
    "duration": 30,
    "output_format": "mp3"
  }
}
```

### Cinematic 8-second cue
Default duration with the stereo-large variant and MP3 output
```json
{
  "model": "meta/musicgen",
  "input": {
    "prompt": "Triumphant cinematic orchestral melody in G major leading to a crescendo",
    "model_version": "stereo-large",
    "duration": 8,
    "output_format": "mp3"
  }
}
```

### 30-second folk melody (large)
Maximum duration with the mono large variant — about 270 credits
```json
{
  "model": "meta/musicgen",
  "input": {
    "prompt": "Acoustic guitar folk melody, gentle and warm",
    "model_version": "large",
    "duration": 30,
    "output_format": "mp3"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Billing is 30.5 + 7.9 credits per second of duration (measured): 8 s ≈ 94, 15 s ≈ 150, 30 s ≈ 270
- Use a melody-* variant with input_audio to keep a melody and change the style
- Set continuation to true to extend an existing clip instead of mimicking it
- multi_band_diffusion only works with the non-stereo variants (melody-large, large)