The image API, in one call.

Generate high-resolution images from text with a single REST request. Predictable JSON, official SDKs, and webhooks for async jobs.

Get your API key Read the quickstart

generate.sh

# Generate an image from a prompt
$ curl https://api.sutertai.com/v1/images \
  -H "Authorization: Bearer $SUTERTAI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-1",
    "prompt": "a black marble monolith on a desert arch, dawn",
    "aspect_ratio": "2:3",
    "resolution": "1248x1872"
  }'

Quickstart

Three steps to your first image.

Create a key

Grab a secret key from your dashboard. Pro and Enterprise plans include API access.

Call `/v1/images`

POST a prompt and parameters. Small images return inline; larger jobs return a job id.

Poll or use a webhook

Fetch /v1/jobs/:id or register a webhook to receive the result when it's ready.

app.py

from sutertai import Sutertai

client = Sutertai(api_key="sk_live_...")

image = client.images.generate(
    model="flux-1",
    prompt="scarlet silk caught mid-motion, ivory",
    aspect_ratio="2:3",
    resolution="1248x1872",
)

# -> https://cdn.sutertai.com/img/9f2c...webp
print(image.url)

The Engine

State-of-the-art models on serious silicon.

Built on leading diffusion architectures, fine-tuned and continuously optimized — then matched to the right GPU for the job.

Stable Diffusion

FLUX

SDXL

A row of GPU server racks with glowing blue LED indicator lights

Dev / Testing

RTX 4090 · 5090

Rapid iteration and experimentation for the research pipeline.

Production · Standard

NVIDIA L4 · L40S

Balanced throughput and quality for everyday generation at scale.

Enterprise · High-scale

H100 · B200

Maximum performance for demanding, high-volume workloads.

Reference

Core endpoints.

POST/v1/imagesGenerate an image from a text prompt

GET/v1/jobs/:idRetrieve the status & result of a job

POST/v1/images/upscaleUpscale an existing image to 2K / 4K

POST/v1/batchesQueue many prompts as a single batch job

GET/v1/batches/:idBatch progress & the finished zip

GET/v1/modelsList available models & capabilities

POST/v1/webhooksRegister a callback for async results

Batch generation

One brief, a whole set of variants.

The Batch Generator on the Product page is this endpoint. Send a list of prompts — or a CSV — and Sutertai queues them as one job, applies the crops you ask for, and returns a single zip when everything is rendered.

batch.py

from sutertai import Sutertai

client = Sutertai(api_key="sk_live_...")

batch = client.batches.create(
    model="flux-1",
    prompts=["a linen tote on warm sand", "the same tote, studio grey"],
    crops=["1:1", "4:5", "9:16"],
    webhook_url="https://yourapp.com/hooks/sutertai",
)

# poll, or wait for the webhook, then download every render at once
done = client.batches.wait(batch.id)
print(done.zip_url)

POST/v1/batchesAccepts prompts[] or a CSV upload, plus optional crops[]

GET/v1/batches/:idReturns progress, per-prompt status and zip_url once complete

Batches count against your monthly image quota, one render per prompt × crop. Available on Pro and Enterprise, same as the rest of the API.

The image API, in one call.

Three steps to your first image.

Create a key

Call `/v1/images`

Poll or use a webhook

State-of-the-art models on serious silicon.

Core endpoints.

One brief, a whole set of variants.

Install and go.

Ship your first call in code.

The image API, in one call.

Three steps to your first image.

Create a key

Call /v1/images

Poll or use a webhook

State-of-the-art models on serious silicon.

Core endpoints.

One brief, a whole set of variants.

Install and go.

Ship your first call in code.

Call `/v1/images`