GPT Image 2 Text-to-Image

OpenAI's GPT Image 2 raw image model can generate high-quality images based on natural language prompts. It provides a ready-to-use REST inference API, offering excellent performance, no cold start, and affordability.

Base URL

https://api.icreat.ai

Authentication

All API requests must be authenticated with an API Key. You can obtain an API Key from the console.

export ICREAT_API_KEY="your-api-key-here"

HTTP Request Headers

import os

API_KEY = os.environ.get("ICREAT_API_KEY")
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer " + API_KEY,
}

Protect your API Key

Never expose your API Key in client-side code or public repositories. Use environment variables or a backend proxy.

Code Examples

Image and video generation uses a three-step async flow: submit to get task_id, poll status, then fetch results when status is SUCCEEDED.

1. Submit Task

Send a generation request to the submit endpoint.

POST/v1/task/submit/openai/gpt-image-2/text-to-image

curl --fail-with-body --connect-timeout 10 --max-time 60 \
  -X POST https://api.icreat.ai/v1/task/submit/openai/gpt-image-2/text-to-image \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
  "prompt": "Replace the background with a city street illuminated by neon lights at night.",
  "image": ["https://cdn.example.com/source/dog.png"],
  "size": "1024x1024",
  "quality": "high"
}'

2. Poll Status

Poll with task_id. Response contains only status.

POST/v1/task/query-status

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/query-status" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
  "task_id": "task-xxx"
}'

3. Get Result

Fetch output when the task succeeds.

POST/v1/task/get-result

curl --connect-timeout 10 --max-time 60 \
  -X POST "https://api.icreat.ai/v1/task/get-result" \
  -H "Authorization: Bearer ${ICREAT_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
  "task_id": "task-xxx"
}'

Input Schema

Submit Task — Input

Total: 4 Required: 3 Optional: 1

promptstringrequired

Text prompt. Maximum length: 32,000 characters.

imagestring[]

A list of reference image URLs. When the list is not empty, image-to-image generation is performed. Supports 1–10 images.

sizestringrequired

The dimensions of the generated image.

qualitystringrequired

Controls how closely the generated result preserves the original input image.

Poll Status — Input

Total: 1 Required: 1 Optional: 0

task_idstringrequired

Task ID from the submit endpoint.

Get Result — Input

Total: 1 Required: 1 Optional: 0

task_idstringrequired

Task ID from the submit endpoint.

Output Schema

Submit Task — Output

Total: 1

task_idstring

Async task identifier.

Poll Status — Output

Total: 1

statusstring

Current task status. Fetch results when SUCCEEDED.

Get Result — Output

Top-level array of resource objects.

Total: 3

typestring

Resource type; Image for this model.

urlstring

Preview/access URL.

download_urlstring

Download URL.

LLM Prompt

The Markdown below is an LLM-friendly prompt you can paste into AI assistants (e.g. Cursor, ChatGPT) to help them understand this model's API, call flow, and key parameters. Use Copy for AI or copy from the code block below.

# openai/gpt-image-2/text-to-image

> OpenAI's GPT Image 2 raw image model can generate high-quality images based on natural language prompts.

## Overview

Use the iCreat three-step async task API: submit a generation request, poll status, then fetch the image URL on success.

## API Info

- **Base URL**：`https://api.icreat.ai`
- **Submit endpoint (POST)**：`/v1/task/submit/openai/gpt-image-2/text-to-image`
- **Poll endpoint (POST)**：`/v1/task/query-status`
- **Get result endpoint (POST)**：`/v1/task/get-result`
- **Model ID**：`openai/gpt-image-2/text-to-image`
- **Auth**：`Authorization: Bearer ${ICREAT_API_KEY}`

## Call Flow

1. **Submit**: POST submit path with body per Input Notes; response `{ "task_id": "..." }`
2. **Poll**: POST `/v1/task/query-status` with `{ "task_id": "..." }`; response contains **only** `{ "status": "SUBMITTED|SUCCEEDED|FAILED" }`
3. **Get result**: when `status` is `SUCCEEDED`, POST `/v1/task/get-result`; top-level array; read `url` or `download_url` when `type` is `Image`

### Input Notes

- Request body uses top-level flat fields
- Required: `prompt`, `size`, `quality`
- Optional: `image`
- `prompt` (required): Text prompt. Maximum length: 32,000 characters.
- `image` (optional): A list of reference image URLs. When the list is not empty, image-to-image generation is performed. Supports 1–10 images.
- `size` (required): The dimensions of the generated image.
- `quality` (required): Controls how closely the generated result preserves the original input image.

### Output Notes

- Poll: read `status` only
- Result: top-level `[{ "type": "Image", "url": "...", "download_url": "..." }]`

## Notes

- Use the same `task_id` across all three steps; do not skip polling
- `FAILED` is terminal — check request parameters or reference media