All models
Qualityflux-2-dev

FLUX.2 Dev

Highest quality FLUX model. Photorealistic, supports image-to-image and inpainting.

Best for: Portraits, photorealism, image-to-image editing, inpainting, premium commercial photography. Slowest but maximum fidelity.

Quality steps
38
Fast preset
25
Max steps
50
Negative prompt
Yes
cURL
POST a single request to generate an image with FLUX.2 Dev.
curl -X POST https://imageapi.org/api/v1/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a cinematic mountain landscape at sunrise",
    "model": "flux-2-dev",
    "width": 1024,
    "height": 1024,
    "num_inference_steps": 38
  }'
Python
import requests

resp = requests.post(
    "https://imageapi.org/api/v1/generate",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "prompt": "a cinematic mountain landscape at sunrise",
        "model": "flux-2-dev",
        "width": 1024,
        "height": 1024,
        "num_inference_steps": 38,
    },
    timeout=120,
)
print(resp.json())
Node.js
const res = await fetch("https://imageapi.org/api/v1/generate", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    prompt: "a cinematic mountain landscape at sunrise",
    model: "flux-2-dev",
    width: 1024,
    height: 1024,
    num_inference_steps: 38,
  }),
});
console.log(await res.json());
Plan availability
FLUX.2 Dev is included on the following plans.
  • pro
  • enterprise

Frequently asked questions

How do I call the FLUX.2 Dev model?
Send a POST request to https://imageapi.org/api/v1/generate with the body { "prompt": "...", "model": "flux-2-dev" } and an Authorization: Bearer YOUR_API_KEY header.
What is FLUX.2 Dev best for?
Portraits, photorealism, image-to-image editing, inpainting, premium commercial photography. Slowest but maximum fidelity.
Does FLUX.2 Dev support negative prompts?
Yes — pass a negative_prompt field to steer FLUX.2 Dev away from unwanted concepts.
What inference step counts work best for FLUX.2 Dev?
Use ~38 steps for production quality and ~25 steps when you want lower latency. The hard cap is 50 steps.
Which plans include FLUX.2 Dev?
FLUX.2 Dev is available on pro, enterprise plans.

Ready to ship with FLUX.2 Dev?

Free tier, transparent pricing, and a single REST endpoint for 9 models.