API Reference

Asynchronous image generation

Gateway asynchronous image extension: create a task with image parameters, then poll for the image. The /v1/videos path does not make this a video task.

POST https://api.tu-zi.com/v1/videos

Preview guide

Paste into an AI assistant to help with integration. Documentation and examples only; no real credentials.

# Asynchronous image generation

Gateway asynchronous image extension: create a task with image parameters, then poll for the image. The /v1/videos path does not make this a video task.

## Endpoint and authentication

`POST https://api.tu-zi.com/v1/videos`

Transport: json

Use an inference API key. Examples contain placeholders; read credentials from environment variables.

```http
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
```

## Body parameters

| Field | Type | Required | Description / constraints |
| --- | --- | --- | --- |
| model | string | Yes | Image model identifier. Example: gpt-image-1 |
| prompt | string | Yes | Image generation prompt. Example: 极简产品摄影，白色陶瓷咖啡杯放在浅色木桌上，柔和自然窗光，干净背景，细节清晰 |
| size | string | No | Output dimensions, for example 1024x1024. Form initial value (not an upstream default): "1024x1024" Values: "1024x1024", "1536x1024", "1024x1536", "512x512", "256x256" Example: 1024x1024 |
| n | number | No | Number of images to generate. Form initial value (not an upstream default): 1 Range in this definition: 1 ~ 10 Example: 1 |
| quality | string | No | Image quality setting. Form initial value (not an upstream default): "standard" Values: "standard", "hd", "auto" Example: standard |
| response_format | string | No | Use url; this gateway forces URL output. Form initial value (not an upstream default): "url" Values: "url", "b64_json" Example: url |
| style | string | No | Image style: vivid or natural, where supported. Values: "vivid", "natural" Example: vivid |
| background | string | No | Background mode; transparent requests transparency where supported. Values: "auto", "transparent", "opaque" Example: auto |

## Minimal request example

```json
{
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}
```

## Before running

Replace https://api.tu-zi.com with this site origin (do not append /v1 twice), and YOUR_MODEL with the exact marketplace model ID. Install requests for Python; use Node.js 20+ and save JavaScript as .mjs. WebSocket samples list their own dependencies.

Create an inference key on the site and set API_KEY in the environment; replace YOUR_API_KEY in cURL. Prepare local input.png/input.wav files for upload examples.

Start with a minimal request, then read the result using the response fields below. For async APIs, a returned task ID requires polling; acceptance does not mean generation has finished.

## Code examples

### curl

```bash
curl --request POST 'https://api.tu-zi.com/v1/videos' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}'
```

### python

```python
import json
import os
import requests

url = "https://api.tu-zi.com/v1/videos"
headers = json.loads("{\"Authorization\":\"Bearer YOUR_API_KEY\",\"Content-Type\":\"application/json\"}")
headers["Authorization"] = "Bearer " + os.environ["API_KEY"]
payload = json.loads("{\"model\":\"YOUR_MODEL\",\"prompt\":\"A white ceramic cup\",\"response_format\":\"url\"}")
response = requests.request("POST", url, headers=headers, json=payload, timeout=120)
response.raise_for_status()
result = response.json()
if isinstance(result, dict) and (result.get("success") is False or result.get("error")):
    raise RuntimeError(json.dumps(result, ensure_ascii=False))
print(result)  # Also inspect protocol-specific status/code fields.
```

### javascript

```javascript
// Node.js 20+
const url = "https://api.tu-zi.com/v1/videos";
const headers = {
  "Authorization": "Bearer YOUR_API_KEY",
  "Content-Type": "application/json"
};
headers.Authorization = 'Bearer ' + process.env.API_KEY;
const response = await fetch(url, { method: "POST", headers, body: JSON.stringify({
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}) });
if (!response.ok) throw new Error(await response.text());
const result = await response.json();
if (result?.success === false || result?.error) throw new Error(JSON.stringify(result));
console.log(result); // Also inspect protocol-specific status/code fields.
```

## Response fields

| Field | Type | Description / constraints |
| --- | --- | --- |
| id / task_id | string | Save the public task ID for polling, not request_id. |
| status | string | Typically queued on submission, not proof of a generated image. |

Illustrative response; some endpoints show essential fields only. Optional fields may be absent.

```json
{
  "id": "task_example",
  "task_id": "task_example",
  "status": "queued"
}
```

## Asynchronous workflow

Save id and GET /v1/videos/{task_id} with the same inference API key. queued and in_progress are pending; completed is terminal success: read video_url and download the image. failed is terminal failure: read error.message and stop. video_url is a compatibility field containing an image URL, not a video; do not parse synchronous data[].url. Bound polling by attempt count and deadline with exponential backoff. Missing result URLs or unknown states require an error/check, not success or endless polling.

The polling endpoint returns JSON; the result URL returns binary image bytes. Never reuse a JSON-only response checker for media. Do not reuse a requests.Session with default Authorization headers to download images: headers={} does not clear session authentication. After polling returns status=completed, use the following helper to download and validate the image. Install Pillow first and use an output_path extension matching the actual image format:

```python
# pip install Pillow
from io import BytesIO
from pathlib import Path
from PIL import Image
import requests

def download_completed_image(task, output_path):
    if task.get("status") != "completed" or not task.get("video_url"):
        raise RuntimeError("Task has not delivered an image URL")
    # A NEW session: do not reuse the authenticated API session.
    with requests.Session() as media_session:
        media_session.trust_env = False
        media = media_session.get(task["video_url"], timeout=(10, 120))
        media.raise_for_status()
        raw = media.content  # Binary image bytes. Never media.json().
    with Image.open(BytesIO(raw)) as image:
        image.verify()
    Path(output_path).write_bytes(raw)
```

## Errors and retries

Check HTTP status and protocol-specific errors; do not mix error formats across protocols. For 401/403 check credentials and permissions, for 429 back off, and for 5xx check gateway/upstream status. Error bodies may contain error/message/code depending on the protocol.

Use bounded exponential backoff for reads. Do not blindly retry task creation, generation, or key mutations, to avoid duplicate billing or execution.

## Constraints

Use only when the current image model explicitly offers this asynchronous endpoint. It shares base image parameters, not video parameters: do not add seconds or duration. Size, quality and count still depend on the model and group. A returned task ID means acceptance only. A usable image URL is required; Base64-only upstreams require enabled conversion and working storage. Never send the site API key when downloading the result URL.

## Documentation source

Definition version: 1

Reviewed: 2026-11-02

Source: https://api.tu-zi.com/en/docs/api/reference-relay-image-generation-async

Related protocol guide: https://api.tu-zi.com/docs/api/images

This export contains documentation and examples only, without real credentials, account data or user requests.

Updated 2026-11-02 · Reviewed 2026-11-02

Endpoint and authentication

POST https://api.tu-zi.com/v1/videos

Transport: json

Use an inference API key. Examples contain placeholders; read credentials from environment variables.

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Minimal request example

{
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}

Before running

Replace https://api.tu-zi.com with this site origin (do not append /v1 twice), and YOUR_MODEL with the exact marketplace model ID. Install requests for Python; use Node.js 20+ and save JavaScript as .mjs. WebSocket samples list their own dependencies.

Create an inference key on the site and set API_KEY in the environment; replace YOUR_API_KEY in cURL. Prepare local input.png/input.wav files for upload examples.

Start with a minimal request, then read the result using the response fields below. For async APIs, a returned task ID requires polling; acceptance does not mean generation has finished.

Constraints

Use only when the current image model explicitly offers this asynchronous endpoint. It shares base image parameters, not video parameters: do not add seconds or duration. Size, quality and count still depend on the model and group. A returned task ID means acceptance only. A usable image URL is required; Base64-only upstreams require enabled conversion and working storage. Never send the site API key when downloading the result URL.

Body parameters

Field	Type	Required	Description / constraints
model	string	Yes	Image model identifier. Example: gpt-image-1
prompt	string	Yes	Image generation prompt. Example: 极简产品摄影，白色陶瓷咖啡杯放在浅色木桌上，柔和自然窗光，干净背景，细节清晰
size	string	No	Output dimensions, for example 1024x1024. Form initial value (not an upstream default): "1024x1024" Values: "1024x1024", "1536x1024", "1024x1536", "512x512", "256x256" Example: 1024x1024
n	number	No	Number of images to generate. Form initial value (not an upstream default): 1 Range in this definition: 1 ~ 10 Example: 1
quality	string	No	Image quality setting. Form initial value (not an upstream default): "standard" Values: "standard", "hd", "auto" Example: standard
response_format	string	No	Use url; this gateway forces URL output. Form initial value (not an upstream default): "url" Values: "url", "b64_json" Example: url
style	string	No	Image style: vivid or natural, where supported. Values: "vivid", "natural" Example: vivid
background	string	No	Background mode; transparent requests transparency where supported. Values: "auto", "transparent", "opaque" Example: auto

Response fields

Field	Type	Description / constraints
id / task_id	string	Save the public task ID for polling, not request_id.
status	string	Typically queued on submission, not proof of a generated image.

Illustrative response; some endpoints show essential fields only. Optional fields may be absent.

{
  "id": "task_example",
  "task_id": "task_example",
  "status": "queued"
}

Asynchronous workflow

Save id and GET /v1/videos/{task_id} with the same inference API key. queued and in_progress are pending; completed is terminal success: read video_url and download the image. failed is terminal failure: read error.message and stop. video_url is a compatibility field containing an image URL, not a video; do not parse synchronous data[].url. Bound polling by attempt count and deadline with exponential backoff. Missing result URLs or unknown states require an error/check, not success or endless polling.

The polling endpoint returns JSON; the result URL returns binary image bytes. Never reuse a JSON-only response checker for media. Do not reuse a requests.Session with default Authorization headers to download images: headers={} does not clear session authentication. After polling returns status=completed, use the following helper to download and validate the image. Install Pillow first and use an output_path extension matching the actual image format:

# pip install Pillow
from io import BytesIO
from pathlib import Path
from PIL import Image
import requests

def download_completed_image(task, output_path):
    if task.get("status") != "completed" or not task.get("video_url"):
        raise RuntimeError("Task has not delivered an image URL")
    # A NEW session: do not reuse the authenticated API session.
    with requests.Session() as media_session:
        media_session.trust_env = False
        media = media_session.get(task["video_url"], timeout=(10, 120))
        media.raise_for_status()
        raw = media.content  # Binary image bytes. Never media.json().
    with Image.open(BytesIO(raw)) as image:
        image.verify()
    Path(output_path).write_bytes(raw)

Errors and retries

Check HTTP status and protocol-specific errors; do not mix error formats across protocols. For 401/403 check credentials and permissions, for 429 back off, and for 5xx check gateway/upstream status. Error bodies may contain error/message/code depending on the protocol.

Use bounded exponential backoff for reads. Do not blindly retry task creation, generation, or key mutations, to avoid duplicate billing or execution.

Document information and sources

Documentation source

Definition version: 1

Reviewed: 2026-11-02

Source: https://api.tu-zi.com/en/docs/api/reference-relay-image-generation-async

Related protocol guide: https://api.tu-zi.com/docs/api/images

This export contains documentation and examples only, without real credentials, account data or user requests.

Request examples

curl --request POST 'https://api.tu-zi.com/v1/videos' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}'

import json
import os
import requests

url = "https://api.tu-zi.com/v1/videos"
headers = json.loads("{\"Authorization\":\"Bearer YOUR_API_KEY\",\"Content-Type\":\"application/json\"}")
headers["Authorization"] = "Bearer " + os.environ["API_KEY"]
payload = json.loads("{\"model\":\"YOUR_MODEL\",\"prompt\":\"A white ceramic cup\",\"response_format\":\"url\"}")
response = requests.request("POST", url, headers=headers, json=payload, timeout=120)
response.raise_for_status()
result = response.json()
if isinstance(result, dict) and (result.get("success") is False or result.get("error")):
    raise RuntimeError(json.dumps(result, ensure_ascii=False))
print(result)  # Also inspect protocol-specific status/code fields.

// Node.js 20+
const url = "https://api.tu-zi.com/v1/videos";
const headers = {
  "Authorization": "Bearer YOUR_API_KEY",
  "Content-Type": "application/json"
};
headers.Authorization = 'Bearer ' + process.env.API_KEY;
const response = await fetch(url, { method: "POST", headers, body: JSON.stringify({
  "model": "YOUR_MODEL",
  "prompt": "A white ceramic cup",
  "response_format": "url"
}) });
if (!response.ok) throw new Error(await response.text());
const result = await response.json();
if (result?.success === false || result?.error) throw new Error(JSON.stringify(result));
console.log(result); // Also inspect protocol-specific status/code fields.