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-2-pro"

"size": "1024x1024"

The live catalog of image models is served by the API. Call GET /v1/models to see everything you can call today and pick a model whose id to pass in the model parameter.

More models are added regularly. See Models Reference for the up-to-date catalog and the Changelog for new additions.

A successful request returns a JSON object with a generation id, a top-level status, and a data[] array containing at least one presigned S3 URL.

{
  "created": 1785845810,
  "id": "6dcc89ce-964e-466e-8100-ba2bdfb6d07c",
  "status": "pending",
  "model": "flux-2-pro",
  "type": "image",
  "data": [
    {
      "url": "https://fluxpool-output.s3.ap-southeast-1.amazonaws.com/...?X-Amz-...",
      "generation_id": "6dcc89ce-964e-466e-8100-ba2bdfb6d07c",
      "status": "pending",
      "model": "flux-2-pro",
      "revised_prompt": null
    }
  ]
}

Field descriptions:

• id — Generation ID. Use this to poll GET /v1/generations/{id}.
• status — One of pending, processing, completed, failed.
• created — Unix timestamp the generation was accepted.
• type — image for this endpoint.
• data[].url — Presigned S3 URL. Becomes fetchable once the generation completes; expires 1 hour after issue.
• data[].revised_prompt — Prompt as revised by the model, if applicable. Usually null.

Image generations on Fluxpool are always asynchronous. The POST returns immediately with a generation id, a top-level status, and a presigned S3 URL that becomes fetchable once the image is ready.

You have two ways to check completion — poll the presigned URL directly, or poll the structured status endpoint.

Option 1 — poll the generation status endpoint:

curl https://api.fluxpool.ai/v1/generations/6dcc89ce-964e-466e-8100-ba2bdfb6d07c \
  -H "Authorization: Bearer fp_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"

Option 2 — poll the presigned S3 URL directly (cheapest with a byte-range GET):

curl -I --range 0-0 "https://fluxpool-output.s3.ap-southeast-1.amazonaws.com/...?X-Amz-..."

Note: S3 presigned GET URLs reject HEAD with a 403. Use GET with Range: bytes=0-0 for a cheap readiness probe.

Status values:

Webhook delivery is on the roadmap — see Async & polling. Until then, poll one of the two endpoints above.

Errors return a JSON object with an error field.

{
  "error": {
    "code": "invalid_model",
    "message": "Model 'flux-9.9-ultra' is not available.",
    "type": "invalid_request_error"
  }
}

All errors include a human-readable message. How many generations can run at once depends on your plan — see Limits.

Complete code examples. Copy, paste, generate.

from openai import OpenAI

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

response = client.images.generate(
    model="flux-2-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_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{"model": "flux-2-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_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"});
const response = await client.images.generate({model: "flux-2-pro", prompt: "A dragon perched on a neon-lit Tokyo rooftop at midnight, cinematic lighting, 8K", size: "1024x1024"});
console.log(response.data[0].url);

response = client.images.generate(
    model="flux-2-pro",
    prompt="Same scene but in cyberpunk style, neon lights, rain",
    size="1024x1024",
    extra_body={"input_images": ["https://your-cdn.com/source-image.jpg"]}
)

import base64

with open("source.png", "rb") as f:
    data_uri = f"data:image/png;base64,{base64.b64encode(f.read()).decode()}"

response = client.images.generate(
    model="flux-2-pro",
    prompt="Same scene but in cyberpunk style, neon lights, rain",
    extra_body={"inputs": [{"type": "image", "source": "data", "value": data_uri}]}
)

• →Video Generation — Generate video with the same API pattern
• →Models Reference — Full parameter support per model
• →Async & polling — How to poll presigned URLs (webhooks coming soon)
• →SDKs & Libraries — Use the OpenAI SDK today
• →Pricing — Per-model generation costs