Image Generation

Requires authentication via Bearer token. See Authentication.

Send a JSON body with the following parameters:

"prompt": "A dragon perched on a neon-lit Tokyo rooftop at midnight, cinematic lighting, 8K"

"model": "flux-1.1-pro"

"size": "1024x1024"

"response_format": "url"

Image generation models available through the endpoint. Prices are per generation at default resolution.

More models are added regularly. See the full list at Models Hub or check the Changelog for new additions.

A successful request returns a JSON object containing the generated image(s).

Response with response_format: "url":

{
  "created": 1719835200,
  "data": [
    {
      "url": "https://cdn.fluxpool.ai/generations/img_abc123.png",
      "revised_prompt": null
    }
  ],
  "model": "flux-1.1-pro",
  "usage": {
    "credits_used": 3,
    "credits_remaining": 47
  }
}

Field descriptions:

• created — Unix timestamp of generation
• data — Array of generated image objects
• data[].url — Temporary URL to the image. Expires in 1 hour. Download promptly.
• data[].revised_prompt — The prompt as modified by the model, if applicable. Most models return null.
• model — The model used
• usage.credits_used — Credits consumed by this generation
• usage.credits_remaining — Your remaining credit balance

Response with response_format: "b64_json":

{
  "created": 1719835200,
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAA..."
    }
  ],
  "model": "flux-1.1-pro",
  "usage": {
    "credits_used": 3,
    "credits_remaining": 47
  }
}

Most image models respond synchronously — the response includes the completed image. For large batches or high-resolution requests, you can use async mode to avoid timeouts.

Set the X-Fluxpool-Async: true header on your request. You'll receive a generation ID immediately. Poll for results or use Webhooks to receive a callback when complete.

Async request:

import requests

response = requests.post(
    "https://api.fluxpool.ai/v1/images/generations",
    headers={
        "Authorization": "Bearer fp_your_api_key",
        "Content-Type": "application/json",
        "X-Fluxpool-Async": "true"
    },
    json={
        "model": "flux-1.1-pro",
        "prompt": "A dragon on a neon-lit Tokyo rooftop, 8K",
        "size": "2048x2048"
    }
)

generation = response.json()
print(generation["id"])       # "gen_xyz789"
print(generation["status"])   # "processing"

Async response:

{
  "id": "gen_xyz789",
  "status": "processing",
  "model": "flux-1.1-pro",
  "created": 1719835200
}

Polling for result:

curl https://api.fluxpool.ai/v1/generations/gen_xyz789 \
  -H "Authorization: Bearer fp_your_api_key"

Completed response:

{
  "id": "gen_xyz789",
  "status": "completed",
  "model": "flux-1.1-pro",
  "created": 1719835200,
  "data": [
    {
      "url": "https://cdn.fluxpool.ai/generations/img_abc123.png"
    }
  ],
  "usage": {
    "credits_used": 5,
    "credits_remaining": 42
  }
}

Status values:

For production pipelines, use Webhooks instead of polling.

Receive a POST callback when an async generation completes. Eliminates polling.

Set the X-Fluxpool-Webhook header to your callback URL:

curl https://api.fluxpool.ai/v1/images/generations \
  -H "Authorization: Bearer fp_your_api_key" \
  -H "Content-Type: application/json" \
  -H "X-Fluxpool-Async: true" \
  -H "X-Fluxpool-Webhook: https://your-app.com/api/webhook" \
  -d '{
    "model": "flux-1.1-pro",
    "prompt": "A dragon on a neon-lit Tokyo rooftop, 8K",
    "size": "2048x2048"
  }'

Webhook payload (delivered to your URL):

{
  "event": "generation.completed",
  "id": "gen_xyz789",
  "status": "completed",
  "model": "flux-1.1-pro",
  "data": [
    {
      "url": "https://cdn.fluxpool.ai/generations/img_abc123.png"
    }
  ],
  "usage": {
    "credits_used": 5
  },
  "signature": "sha256=a1b2c3d4..."
}

Verify the signature field to confirm the webhook is from Fluxpool. See Webhooks documentation for signature verification details and retry logic.

Errors return a JSON object with an error field.

{
  "error": {
    "code": "invalid_model",
    "message": "Model 'flux-9.9-ultra' is not available. See /docs/models for supported models.",
    "type": "invalid_request_error"
  }
}

Common errors:

All errors include a human-readable message. Log these for debugging. Rate limits vary by plan — see Authentication for details.

Complete code examples. Copy, paste, generate.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.fluxpool.ai/v1",
    api_key="fp_your_api_key"
)

response = client.images.generate(
    model="flux-1.1-pro",
    prompt="A dragon perched on a neon-lit Tokyo rooftop at midnight, cinematic lighting, 8K",
    size="1024x1024"
)

print(response.data[0].url)

curl https://api.fluxpool.ai/v1/images/generations \
  -H "Authorization: Bearer fp_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-1.1-pro",
    "prompt": "A dragon perched on a neon-lit Tokyo rooftop at midnight, cinematic lighting, 8K",
    "size": "1024x1024"
  }'

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.fluxpool.ai/v1",
  apiKey: "fp_your_api_key",
});

const response = await client.images.generate({
  model: "flux-1.1-pro",
  prompt: "A dragon perched on a neon-lit Tokyo rooftop at midnight, cinematic lighting, 8K",
  size: "1024x1024",
});

console.log(response.data[0].url);

# Qwen — great for Asian aesthetics and multilingual prompts
response = client.images.generate(
    model="qwen-image-3-6-plus",
    prompt="水墨画风格的山水画，云雾缭绕的山峰，传统中国画",
    size="1024x1536"
)

# SDXL — fast and cheap, ControlNet compatible
response = client.images.generate(
    model="sdxl-1.0",
    prompt="Isometric RPG village with tavern and market, pixel art style",
    size="1024x1024"
)

# Ideogram — best text rendering in images
response = client.images.generate(
    model="ideogram-2.0",
    prompt='A neon sign that reads "OPEN 24/7" on a rainy city street',
    size="1024x1024"
)

response = client.images.generate(
    model="flux-1.1-pro",
    prompt="Product photo of white sneakers on marble surface, studio lighting, clean background",
    size="1536x1024",  # Landscape
)

# With advanced parameters (pass via extra_body)
response = client.images.generate(
    model="sdxl-1.0",
    prompt="Fantasy castle on a cliff at sunset, epic cinematic lighting",
    size="1024x1024",
    extra_body={
        "negative_prompt": "blurry, low quality, distorted, watermark",
        "steps": 40,
        "guidance_scale": 7.5,
        "seed": 42
    }
)

curl https://api.fluxpool.ai/v1/images/generations \
  -H "Authorization: Bearer fp_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "sdxl-1.0",
    "prompt": "Professional portrait photo of a woman, natural lighting",
    "size": "1024x1024",
    "negative_prompt": "blurry, distorted, extra limbs, watermark, text, low quality"
  }'

response = client.images.generate(
    model="sdxl-1.0",
    prompt="Concept art for a sci-fi weapon, clean design, white background",
    size="1024x1024",
    n=4
)

for i, image in enumerate(response.data):
    print(f"Image {i+1}: {image.url}")

response = client.images.generate(
    model="sdxl-1.0",
    prompt="Same scene but in cyberpunk style, neon lights, rain",
    size="1024x1024",
    extra_body={
        "image": "https://example.com/your-source-image.jpg",
        "strength": 0.7  # 0.0 = keep original, 1.0 = ignore original
    }
)

• → Video Generation — Generate video with the same API pattern
• → Models Reference — Full parameter support per model
• → Webhooks — Receive callbacks for async generations
• → SDKs & Libraries — Python, Node.js, and OpenAI SDK usage
• → Pricing — Per-model generation costs
• → Playground — Try image generation in the browser