ComfyUI Node
SparkDiffusion Image To Video
Image-to-video with the official SparkDiffusion Wan2.1 I2V inference in an isolated runtime.
SparkDiffusion Image To Video
- runtime
- profile
- image
- video
- frames
- video_path
- metadata
◄promptThe subject slowly turns toward the camera.►
◄seed0►
◄num_frames81►
◄filename_prefixsparkdiffusion/SparkWanI2V►
◄fixed_resolutionfalse►
◄load_framesfalse►
◄num_videos1►
CategorySparkDiffusion
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| runtime | SPARKDIFFUSION_RUNTIME | — | |
| profile | SPARKDIFFUSION_PROFILE | — | |
| image | IMAGE | — | |
| prompt | STRING | The subject slowly turns toward the camera. | — |
| seed | INT | 00–4294967295 | — |
| num_frames | INT | 815–161 | — |
| filename_prefix | STRING | sparkdiffusion/SparkWanI2V | — |
| fixed_resolutionopt | BOOLEAN | false | Off = upstream input-aspect sizing (keeps the image aspect at the profile's pixel area). On = force the profile resolution at 16:9. |
| load_framesopt | BOOLEAN | false | — |
| num_videosopt | INT | 11–16 | Videos generated in ONE runtime process with seeds seed, seed+1, ... (upstream NUM_SAMPLES). Loading and torch.compile are paid once, so extra videos are cheap. |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |
| frames | IMAGE | — |
| video_path | STRING | — |
| metadata | STRING | — |