# FLUX Dev - Core.Today AI API

> A 12 billion parameter rectified flow transformer capable of generating images from text descriptions, tuned for open, high-quality experimentation.

- **Provider**: Black Forest Labs
- **Model ID**: black-forest-labs/flux-dev
- **Category**: Image Generation
- **Credits**: 58 per call (flat, regardless of num_outputs 1-4)
- **Speed**: Fast
- **Quality**: High

## Features
- 12B parameter rectified flow transformer
- Optional fp8 'go_fast' mode for faster generation
- Supports image-to-image via the image parameter
- Adjustable inference steps (1-50) and guidance
- Up to 4 outputs per call

## Use Cases
- Rapid prototyping and experimentation with prompts
- Style exploration before committing to a premium model
- Image-to-image restyling from a reference photo
- Batch concept art generation

## 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 - Prompt for generated image

### Optional
- **aspect_ratio**: string (default: 1:1) - Aspect ratio for the generated image
  Options: 1:1, 16:9, 21:9, 3:2, 2:3, 4:5, 5:4, 3:4, 4:3, 9:16, 9:21
- **image**: string - Input image for image-to-image mode. Output aspect ratio matches this image
- **prompt_strength**: number (default: 0.8) - Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
- **num_outputs**: integer (default: 1) - Number of outputs to generate (1-4). Billed at the same flat rate regardless of count
- **num_inference_steps**: integer (default: 28) - Number of denoising steps (recommended 28-50). Fewer steps = faster, lower quality
- **guidance**: number (default: 3) - Guidance scale for generated image (0-10)
- **seed**: integer - Random seed. Set for reproducible generation
- **output_format**: string (default: webp) - Format of the output images
  Options: webp, jpg, png
- **output_quality**: integer (default: 80) - Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs
- **disable_safety_checker**: boolean (default: false) - Disable safety checker for generated images.
- **go_fast**: boolean (default: true) - Run faster fp8-quantized predictions; disable for original bf16 (non-deterministic when enabled)
- **megapixels**: string (default: 1) - Approximate number of megapixels for generated image
  Options: 1, 0.25

## Examples

### Text-to-Image (Default)
Generate an image from a text prompt with default settings
```json
{
  "model": "black-forest-labs/flux-dev",
  "input": {
    "prompt": "black forest gateau cake spelling out the words \"FLUX DEV\", tasty, food photography, dynamic shot",
    "aspect_ratio": "1:1",
    "go_fast": true
  }
}
```

## 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
- num_outputs (1-4) is billed at the same flat rate — request multiple outputs in one call to save credits over separate calls
- Use go_fast=true for quick iteration, then disable it for a final bf16 render
- Supply image for image-to-image editing; the output aspect ratio will match the input
- Lower num_inference_steps for fast drafts, raise it towards 50 for final quality

## Documentation
https://replicate.com/black-forest-labs/flux-dev