[API reference]

Images

Generate images from a text prompt with GPT Image 2 or FLUX.2 [pro]. Images are priced per image and returned as URLs or base64.

POST/v1/images/generations

Generate an image

curl https://api.topolabs.ai/v1/images/generations \  -H "Authorization: Bearer $TOPO_API_KEY" \  -H "Content-Type: application/json" \  -d '{    "model": "flux.2-pro",    "prompt": "Topographic contour map at night, thin lime lines on black",    "size": "1536x1024"  }'

image = client.images.generate(    model="flux.2-pro",    prompt="Topographic contour map at night, thin lime lines on black",    size="1536x1024",) print(image.data[0].url)

const image = await client.images.generate({  model: "flux.2-pro",  prompt: "Topographic contour map at night, thin lime lines on black",  size: "1536x1024",}); console.log(image.data[0].url);

Request body

modelstringRequired
gpt-image-2 or flux.2-pro.
promptstringRequired
What to draw. Up to 4,000 characters.
ninteger
Number of images, from 1 to 4. Each is billed separately.
Default: 1
sizestring
1024x1024, 1536x1024 or 1024x1536.
Default: 1024x1024
response_formatstring
url or b64_json.
Default: url
seedinteger
Fix the seed for repeatable output. FLUX models only.

Response

response.json

json

{  "created": 1792142596,  "data": [{ "url": "https://cdn.topolabs.ai/img/7Hq2kf91.png" }],  "topo": { "node": "us-west-2", "gpu": "H100", "receipt": { "amount": "0.0300", "currency": "USDT" } }}

Image models

Model	Kind	Price
GPT Image 2 `gpt-image-2`· OpenAI	Image Per image	$0.053 / image
FLUX.2 [pro] `flux.2-pro`· Black Forest Labs	Image Per image	$0.03 / image
Veo 3.1 `veo-3.1`· Google	Video Per second	$0.40 / second

Video

Video models such as veo-3.1 are billed per second of output ($0.40 per second) and run as async jobs on POST /v1/videos/generations. Poll the job, or pass a webhook_url to be called when it finishes.