Image Generation

Generate images with the Oxen.ai API. Each request returns a generation ID right away, and you poll it until the image is ready, usually in under 30 seconds.

Generate an image

Python
cURL

import os
import requests
import time

HEADERS = {
    "Authorization": f"Bearer {os.environ['OXEN_API_KEY']}",
    "Content-Type": "application/json",
}

enqueue = requests.post(
    "https://hub.oxen.ai/api/ai/queue",
    headers=HEADERS,
    json={
        "model": "gpt-image-2-5-sunburst",
        "prompt": "A red cube on a white background, minimal",
    },
)
enqueue.raise_for_status()
generation_id = enqueue.json()["generations"][0]["generation_id"]
print(f"Enqueued {generation_id}")

while True:
    data = requests.get(
        f"https://hub.oxen.ai/api/ai/queue/{generation_id}",
        headers=HEADERS,
    ).json()
    print(f"  {data['status']}")
    if data["status"] in {"succeeded", "failed", "cancelled"}:
        break
    time.sleep(3)

if data["status"] != "succeeded":
    raise SystemExit(f"{data['status']}: {data['error_message']}")

with open("output.png", "wb") as f:
    f.write(requests.get(data["result_url"]).content)
print("Saved output.png")

Start the generation

GEN_ID=$(curl -s -X POST https://hub.oxen.ai/api/ai/queue \
  -H "Authorization: Bearer $OXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-5-sunburst",
    "prompt": "A red cube on a white background, minimal"
  }' | jq -r '.generations[0].generation_id')

Wait for it to finish

while true; do
  STATUS=$(curl -s -H "Authorization: Bearer $OXEN_API_KEY" \
    "https://hub.oxen.ai/api/ai/queue/$GEN_ID" | jq -r '.status')
  echo "$STATUS"
  case $STATUS in succeeded|failed|cancelled) break;; esac
  sleep 3
done

Download the result

curl -s -H "Authorization: Bearer $OXEN_API_KEY" \
  "https://hub.oxen.ai/api/ai/queue/$GEN_ID" | jq -r '.result_url' \
  | xargs curl -o output.png

Start from an image

Add input_image to start from an image. Editing models like qwen-image-edit change it, and many generation models use it as a reference.

import os
import requests

response = requests.post(
    "https://hub.oxen.ai/api/ai/queue",
    headers={
        "Authorization": f"Bearer {os.environ['OXEN_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": "gpt-image-2-5-sunburst",
        "prompt": "Make the background blue",
        "input_image": ["https://hub.oxen.ai/api/repos/elau/assets/file/main/bloxy/bloxy_cropped_512x512.png"],
    },
)

print(response.json())

curl -X POST https://hub.oxen.ai/api/ai/queue \
  -H "Authorization: Bearer $OXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-5-sunburst",
    "prompt": "Make the background blue",
    "input_image": ["https://hub.oxen.ai/api/repos/elau/assets/file/main/bloxy/bloxy_cropped_512x512.png"]
  }'

The URLs must be public, since the model downloads them.

Generate several at once

Set num_generations to get up to 4 images from one request. Each has its own generation ID.

curl -X POST https://hub.oxen.ai/api/ai/queue \
  -H "Authorization: Bearer $OXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-5-sunburst",
    "prompt": "Abstract geometric pattern in blue and gold",
    "num_generations": 4
  }'

Inference API

Image Generation

Generate an image

Start from an image

Generate several at once

What’s next

​Generate an image

​Start from an image

​Generate several at once

​What’s next

Generate an image

Start from an image

Generate several at once

What’s next