Qwen Video Model
Qwen videos are classified into a video model directory, suitable for video generation, video extension, and model compatibility with request. MoonNexAI connects Qwen video to a unified video task process, which is then submitted with task ID to query status and results.
Current Model
Qwen video model names are based on MoonNexAI console, model lists API and API Reference. It is recommended to call GET /v1/models to confirm the available model before access.
Current API Behaviour
| Item | Current behaviour |
|---|---|
| Create task | Use POST /v1/videos. |
| Video Extension | Use POST /v1/videos/extend for support. |
| Query task | Use GET /v1/videos/{task_id}, or you can also search by following the task central process. |
| Enter Recommendations | Prompt writes about the subject, the action, the lens, the style and the content that needs to be continued. |
API Path
| Methodology | Path | Purpose |
|---|---|---|
POST | /v1/videos | Creates a Qwen video task. |
POST | /v1/videos/extend | Submit video extension task. |
GET | /v1/videos/{task_id} | Query video task results. |
GET | /v1/tasks/{task_id} | Query task centre state. |
Common field
| field | Annotations |
|---|---|
model | Qwen Video Model Name, based on the Model List and API Reference. |
prompt | Video generation or extension description. |
duration | Video duration. |
aspect_ratio | Video aspect ratio. |
image / image_url | image-to-video or reference images input. |
video / video_url | reference video is imported when the video is extended. |
callback_url | task addresses after completion of callback. |
Example request
bash
curl https://moonnexai.com/v1/videos \
-H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen-video-model",
"prompt": "一段产品宣传视频，主体居中，镜头缓慢推进，真实棚拍光照。",
"duration": 5,
"aspect_ratio": "16:9"
}'task Processing
| Phase | Treatment |
|---|---|
| Successful submission | Saves task ID, model name, prompt and input media assets. |
| Generating | polling query API, or wait for callback notification. |
| Completed | Reads the video URL and saves the full query result. |
| Failed | Saves error information and request parameters to facilitate resubmission. |