ComfyUI Node
Modelsite Text to Video
Generate a video from a text prompt with a model from the Modelsite catalog.
Modelsite Text to Video
- video
- task_id
- usage
◄modeldreamina-seedance-2-0►
◄prompt►
◄resolution720p►
◄ratio16:9►
◄duration5►
◄seed-1►
◄generate_audiotrue►
◄watermarkfalse►
◄negative_prompt►
◄timeout_minutes15►
◄custom_model►
CategoryModelsite/Video
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | dreamina-seedance-2-0 | Models are fetched from the Modelsite catalog when ComfyUI loads this page. Refresh the page to update the list. |
| prompt | STRING | What the video should show. English or Chinese. | |
| resolution | COMBO | 720p | Wan and HappyHorse accept 720p or 1080p only; Seedance 480p/720p/1080p; MiniMax H3 768p (H3 Max also 480p). |
| ratio | COMBO | 16:9 | 6 options: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 |
| duration | INT | 52–15 | Seconds. Seedance 4-15, Wan 2-15, HappyHorse 3-15, MiniMax per model. |
| seed | INT | -1-1–2147483647 | -1 lets the model choose. |
| generate_audio | BOOLEAN | true | Seedance only; other models ignore it. |
| watermark | BOOLEAN | false | — |
| negative_promptopt | STRING | Wan and HappyHorse only; other models ignore it. | |
| timeout_minutesopt | INT | 151–60 | How long to wait before giving up. A task that times out is NOT re-created and may still be billed. |
| custom_modelopt | STRING | Optional. A model id that is not in the list yet (for example a model added after you loaded this page). Overrides 'model'. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |
| task_id | STRING | — |
| usage | STRING | — |