Overview of the video series
Video series is used to create text-to-video, image-to-video, reference media assets, video extensions and Remix. Video task is usually asynchronous tasks: returns task ID after submission and obtains the final result by searching API or WebHOK.
Matrix Video Model Matrix Select an endpoint by model, input references, duration, aspect ratio and query method.POSTCreate video tasksSubmit text-to-video, image-to-video and reference-based video tasks.GETQuery video tasksRead status, result URLs, failure details and troubleshooting fields.UploadsUploads and assetsCheck when to supply URLs directly and when hosted assets or asset references are needed.
Route List
| Methodology | Path | Purpose |
|---|---|---|
POST | /v1/videos | The recommended video task creates the portal. |
POST | /v1/video/generations | Compatible video task creates access. |
GET | /v1/tasks/{task_id} | Recommended unified task query portal. |
GET | /v1/videos/{task_id} | A compatible video task query portal. |
GET | /v1/videos/{task_id}/content | Can not get folder: %s: %s |
POST | /v1/videos/{video_id}/remix | Video Remix. |
POST | /v1/videos/extend | Video extension. |
POST | /sora/v1/characters | Creates a video role. |
Basic processes
- View Video Model Matrix to confirm the input, aspect ratio and duration support model.
- If a stable public network media assets URL exists, priority is given to the URL being passed directly in the generation of request.
- Read Uploading and Media Asset Overview first when MoonNexAI is required to host URL or media assets references.
- Call Create Video Job, save returned
idortask_id. - Call Query Video Job, or wait for Webhook.
Model Directory
| Contents | Fit to scene |
|---|---|
| Sora Video | text-to-video, reference images video, role reuse and Remix. |
| Veo Video | High quality text-to-video, image-to-video and reference media assets videos. |
| Seedance-2 Video | SZ, Original, KZ, CL, Feedance 2.5 CL, CL-B, GL series, using their respective public model names, parameters and media assets processes. |
| Moon-2 Video | The DJ, WC, GL, LD, AM, IL series, which is produced in seconds and multi-media assets by sub-generation. |
| HappyHorse / Wan video | Text-to-video, image-to-video and reference-based generation with happyhorse-1.0 and wan2.7. |
| Grok video | Single-image generation with grok-video-1.5 and multiple reference images with grok-image-video. |
| Qwen Video | Video generation and video extension under other video models. |
Standard request
bash
curl https://moonnexai.com/v1/videos \
-H "Authorization: Bearer <MOONNEXAI_API_KEY>" \
-H "Content-Type: application/json" \
-d '{
"model": "seedance-2.0-kz-fast",
"prompt": "A cinematic close-up of a glass perfume bottle on a reflective table.",
"duration": 5,
"aspect_ratio": "16:9",
"callback_url": "https://example.com/moonnexai/webhook"
}'