ComfyUI Node
FL_MiniMaxH3MotionPrepare
A ComfyUI node in FL/MiniMax H3/Internal with 12 inputs and 1 output.
FL_MiniMaxH3MotionPrepare
- vae
- positive
- latent
- audio_vae
- baseline_images
- baseline_audio
- shot_data
- FL_H3_MOTION_PREP
◄target_long_side0►
◄motion_coveragebalanced►
◄max_hold4►
◄expand_to_endtrue►
◄upscale_methodlanczos►
CategoryFL/MiniMax H3/Internal
Inputs (12)
| Name | Type | Default | Description |
|---|---|---|---|
| vae | VAE | — | |
| target_long_side | INT | 00–16384 | 0 keeps source size. Otherwise upscale the long edge; dimensions align to 32 pixels. |
| motion_coverage | COMBO | balanced | Off benchmarks spatial refinement without temporal stretching. Uniform uses max_hold everywhere. |
| max_hold | INT | 42–8 | Used only by uniform coverage. Adaptive coverage presets own their hold limits. |
| expand_to_end | BOOLEAN | true | Extend a short unheld tail through the last motion burst, matching H3 Time Smear. |
| upscale_method | COMBO | lanczos | 3 options: lanczos, bicubic, bilinear |
| positiveopt | CONDITIONING | Required without shot_plan. Shot-plan mode uses each shot's own conditioning instead. | |
| latentopt | LATENT | Completed native H3 video/audio latent, not an empty latent or temporal reshot. | |
| audio_vaeopt | VAE | Required when the timeline expands; not needed with coverage off. | |
| baseline_imagesopt | IMAGE | Optional shared decode of this exact source latent. Connect the same images used by the before panel. | |
| baseline_audioopt | AUDIO | Optional shared decode of this exact source latent's soundtrack. | |
| shot_dataopt | FL_H3_MOTION_SHOT | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| FL_H3_MOTION_PREP | FL_H3_MOTION_PREP | — |