Images

Generate

POST /v1/images/generations

{
  "model": "flux-schnell",
  "prompt": "a clean product photo of a matte black coffee mug",
  "n": 1,
  "size": "1024x1024"
}

Some providers return immediately with HTTP 200. Async providers return HTTP 202 with a task_id.

Edit

POST /v1/images/edits

Use multipart/form-data with an image file and a text prompt.

curl https://api.parel.cloud/v1/images/edits \
  -H "Authorization: Bearer pk-dev-YOUR_KEY" \
  -F model="gpt-image-1.5" \
  -F prompt="Change the mug color to white" \
  -F image="@mug.png"

Image operations

Endpoint	Purpose
POST /v1/images/upscale	Upscale an image
POST /v1/images/bg-remove	Remove an image background

Async polling

import time
import httpx

headers = {"Authorization": "Bearer pk-dev-YOUR_KEY"}
resp = httpx.post(
    "https://api.parel.cloud/v1/images/generations",
    headers=headers,
    json={"model": "flux-schnell", "prompt": "a cat in space"},
)

if resp.status_code == 202:
    task_id = resp.json()["task_id"]
    while True:
        task = httpx.get(f"https://api.parel.cloud/v1/tasks/{task_id}", headers=headers).json()
        if task["status"] == "completed":
            print(task["result"]["data"][0]["url"])
            break
        if task["status"] == "failed":
            raise RuntimeError(task["error"]["message"])
        time.sleep(3)