GPT Image 2.6 API
Prepare a visual brief with GPT Image 2.6. Describe a new image or changes to reference images, then review aspect ratio, resolution, quality, and file settings in the APIMart workspace.
GPT Image 2.6 API configuration
- Prompts and optional reference images
- Aspect ratio, resolution, and quality
- PNG, JPEG, or WebP output settings
Describe the subject, composition, lighting, and any text to include.
Transparent backgrounds require PNG or WebP.
Reference images
Add up to 16 images with public HTTP(S) URLs or local PNG, JPEG, and WebP files (up to 10 MB each).
Your generated images will appear here
Plan your next image with GPT Image 2.6 API
Turn a rough idea into a specific image brief. Keep the subject, composition, and output settings together before reviewing the request configuration.
GPT Image 2.6 API examples: text to image and image editing
These examples use APIMart’s common asynchronous image-task format. Review the model value and workspace options, then follow the task-ID and result-query structure.
- 1
Authenticate the GPT Image 2.6 API example
Save an APIMart key as APIMART_API_KEY. Use Authorization: Bearer <key> for the submission and task-query requests.
- 2
Build the image request
The example sends POST /v1/images/generations with model set to gpt-image-2.6, prompt, and output settings. For editing, add image_urls. The common task response stores the ID in data[0].task_id.
- 3
Query the task result
Use GET /v1/tasks/{task_id}. For a completed task, read data.result.images[0].url[0]; for a failed task, read data.error.message.
See the APIMart documentation for APIMart authentication and general API usage.
curl -X POST https://api.apimart.ai/v1/images/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.6",
"prompt": "A ceramic coffee cup on a pale wooden table, soft window light, ample empty space on the right, editorial product photograph.",
"size": "1:1",
"resolution": "1k",
"quality": "auto",
"n": 1,
"output_format": "png",
"background": "auto"
}'import os
import time
import requests
API_BASE = "https://api.apimart.ai"
headers = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}"}
# 1. Submit the generation task
payload = {
"model": "gpt-image-2.6",
"prompt": "A ceramic coffee cup on a pale wooden table, soft window light, ample empty space on the right, editorial product photograph.",
"size": "1:1",
"resolution": "1k",
"quality": "auto",
"n": 1,
"output_format": "png",
"background": "auto"
}
resp = requests.post(f"{API_BASE}/v1/images/generations", headers=headers, json=payload, timeout=60)
resp.raise_for_status()
task_id = resp.json()["data"][0]["task_id"]
# 2. Poll the task until it finishes
for _ in range(200):
task_resp = requests.get(f"{API_BASE}/v1/tasks/{task_id}", headers=headers, timeout=30)
task_resp.raise_for_status()
task = task_resp.json()["data"]
if task["status"] == "completed":
print(task["result"]["images"][0]["url"][0])
break
if task["status"] == "failed":
raise RuntimeError(task.get("error", {}).get("message", "generation failed"))
time.sleep(3)
else:
raise TimeoutError(f"task {task_id} did not finish in time; continue GET polling with this task ID")curl -X POST https://api.apimart.ai/v1/images/generations \
-H "Authorization: Bearer $APIMART_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-image-2.6",
"prompt": "Move the product in the reference image onto a light stone countertop, keeping its shape, colors, and label.",
"size": "auto",
"resolution": "1k",
"quality": "auto",
"n": 1,
"output_format": "png",
"background": "auto",
"image_urls": [
"https://example.com/product.png"
]
}'TASK_ID="task_xxx" # replace with the task_id returned by the POST request
curl "https://api.apimart.ai/v1/tasks/$TASK_ID" \
-H "Authorization: Bearer $APIMART_API_KEY"GPT Image 2.6 API parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
model | string | Required | Default: gpt-image-2.6 | The model value used by this workspace and its examples: gpt-image-2.6. |
prompt | string | Required | Default: — | Describe the image to create, or the changes to make and the reference details to preserve. |
size | string | Optional | Default: auto | Workspace choices: auto, 1:1, 3:2, 2:3, 16:9, 9:16, 4:3, 3:4, 4:5, or 5:4. |
resolution | string | Optional | Default: 1k | Workspace resolution choices: 1k, 2k, or 4k. This is a request setting, not a claim about native output dimensions. |
quality | string | Optional | Default: auto | Workspace quality choices: auto, low, medium, or high. |
n | integer | Optional | Default: 1 | Image count in this workspace: an integer from 1 to 4. |
output_format | string | Optional | Default: png | Output-format choices: png, jpeg, or webp. Use png or webp with a transparent background. |
background | string | Optional | Default: auto | Background choices: auto, opaque, or transparent. Transparent cannot be combined with jpeg. |
image_urls | array | Optional | Default: — | Optional array of up to 16 HTTP or HTTPS reference-image URLs in this workspace. Local files are for local preview; use reachable image URLs in request examples. |
