Source: 创建AI任务 · Category: Interface — AI Image Generation (接口-AI生图) · Updated 2026-09-22 16:32:14
English Translation
Create AI Task
Submit synthesis and receive the image — save the task ID and receive the image result according to the callback description in this document.
What to prepare before starting?
Prepare your merchant ID, merchant secret key, and the image-generation callback URL. AI uses merchant credentials, not the printer-device key.
AI tasks are asynchronous: after the task ID is returned, wait for the callback to receive the synthesized image, then submit a print task separately. A successful image generation does not automatically complete printing.
Third-party platforms call the Create AI Task interface to submit AI image-generation requests to the Open Platform. The platform synchronously returns the AI task ID (
prompt_id) and automatically dispatches it to the device terminal software. Task creation is supported even when the device is offline; once the device comes online it can view and execute the pending generation/print tasks.
Interface Information
| Item | Value |
|---|---|
| Endpoint | http://www.colorpark.cn/api/ai/run |
| Method | POST |
Request Parameters
| Parameter | Example Value | Required | Description |
|---|---|---|---|
partner_id |
2025000001 |
Yes | Merchant ID provided by the Open Platform (view in the Open Platform: Merchant Management → My Merchant → Number) |
once |
8305 |
Yes | Random value used when generating the encrypted signature |
time |
1646790335 |
Yes | Timestamp used when generating the encrypted signature |
sign |
9ce3ce7a3fccaf3d7715bd7e9bb1dc1c |
Yes | Signature string obtained from the "Generate Merchant Signature" interface |
image |
http://rfccrm.oss-cn-shenzhen.aliyuncs.com/temp/502707bf1b65e307c43bcb86aa10746a43107961.png |
Yes | Original image URL (the oss_image from the template) |
template_id |
39 |
Yes | 39 — dual-image (image = uploaded image, scene = scene image); 40 — single-image (image = uploaded image) |
scene |
http://rfccrm.oss-cn-shenzhen.aliyuncs.com/render/2025/11/26/1124014/2d1852e24c659bbaeeca4e44e972426e7448755f.png |
Yes | Scene image URL |
callback |
http://www.abc.com/api/callback |
Yes | Callback URL — the address used by the client's system to receive task-submission notifications |
prompt |
Replace the person in image 1 completely onto the person's position in image 2, keeping image 2's original background, lighting, perspective, and tone unchanged; match the figure's size ratio to the original, blend naturally with no dissonance, realistic details, high-quality compositing, HD, 8K, rich detail, clean edges, unified lighting, natural fusion, professional-grade compositing, flawless. |
Yes | Prompt (in Chinese) |
comment |
合照 (Group photo) |
No | Third-party style name or remark |
app_number |
10908002 |
No | Application number of the AI photo device |
image_out |
{"angle":45,"x_scale":1.5,"y_scale":1.5} |
No | Image expansion. Rotate image: {"angle":45,"x_scale":1.5,"y_scale":1.5}; proportional image: {"x_scale":2,"y_scale":2,"best_quality":false,"limit_image_size":true}; expand by specified direction: {"left_offset":200,"right_offset":100,"best_quality":false,"limit_image_size":true}; expand by specified aspect ratio: {"angle":0,"output_ratio":"4:3","best_quality":false,"limit_image_size":true} |
width |
— | No | Image width, default 1440px |
height |
— | No | Image height, default 2560px |
Recommended Width/Height Pixel Values
| Resolution | Aspect Ratio | Width × Height (px) |
|---|---|---|
| 2K | 1:1 | 2048 × 2048 |
| 2K | 4:3 | 2304 × 1728 |
| 2K | 3:4 | 1728 × 2304 |
| 2K | 16:9 | 2848 × 1600 |
| 2K | 9:16 | 1600 × 2848 |
| 2K | 3:2 | 2496 × 1664 |
| 2K | 2:3 | 1664 × 2496 |
| 2K | 21:9 | 3136 × 1344 |
| 3K | 1:1 | 3072 × 3072 |
| 3K | 4:3 | 3456 × 2592 |
| 3K | 3:4 | 2592 × 3456 |
| 3K | 16:9 | 4096 × 2304 |
| 3K | 9:16 | 2304 × 4096 |
| 3K | 2:3 | 2496 × 3744 |
| 3K | 3:2 | 3744 × 2496 |
| 3K | 21:9 | 4704 × 2016 |
Callback Notification
| Item | Value |
|---|---|
| Method | POST |
| Headers | Content-Type: application/x-www-form-urlencoded |
Failure example:
{"status":"fail","prompt_id":10449,"picture":"","message":"Image type error, please check the image type and content"}
Success example:
{"status":"success","prompt_id":11047,"picture":"https://rfccrm.colorpark.cn/colorcloud/ai/406d414e7167a4b09bae92626e3fed6585980d11.png","message":"generated successfully"}
Note: The source success example had a malformed/concatenated URL and trailing
"generated/successfullyfragments; normalized above to a valid JSON string.
Tips:
template_id=17: AI image expansion — thetextparameter may be empty.- If the model is portrait style, the
style_indexattribute must be passed; otherwise it need not be passed. - Creating an AI task is an asynchronous task; image generation is notified via an asynchronous callback through the
callbackattribute. Async callback result — headersContent-Type=application/x-www-form-urlencoded, form params:status="success",picture="http://abc.com/xxx.jpeg",prompt_id="xxxxxxxx",message="".
Callback field meanings:
| Field | Meaning |
|---|---|
status |
Status: queueing — in queue; success — succeeded; fail — failed; cancel — pending cancellation; canceled — canceled |
picture |
Image URL |
prompt_id |
Task ID |
message |
Information hint |
Response Result
Note: The source table header repeats "参数名 参数名"; normalized below as Parameter / Example Value / Description.
| Parameter | Example Value | Description |
|---|---|---|
code |
0 |
Result status code — 0 on success, 400 on failure |
msg |
请求成功 (Request successful) |
Result description, e.g. "Request successful" |
data.prompt_id |
c20d543a-8c8f-4b55-81da-3c54a5ba93ca |
AI task ID |
If you encounter unsolvable issues while debugging the device, contact the cloud-printing representative for troubleshooting. Contact details: Tel: 137-5177-4454 [WeChat same number] Email: info@ColorPark.cn