ComfyUI Node
[Legacy] H3 Continuum Sampler V3
Latent-first N-chunk MiniMax H3 sampler. Sampling, native continuation, State/Session, and Spectrum interop stay inside Continuum; ComfyUI Core performs video and audio VAE decoding.
[Legacy] H3 Continuum Sampler V3
- model
- clip
- video_vae
- sampler
- sigmas
- first_frame
- prompt_overrides
- advanced
- video_latents
- audio_latents
- assembly_plan
- result
◄sequence_prompt—►
◄prompt_modeAuto►
◄chunks3►
◄chunk_seconds5.0►
◄width1344►
◄height768►
◄continuityBalanced — 22 frames►
◄base_seed0►
CategoryMiniMax H3/Continuum/Legacy
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| clip | CLIP | — | |
| video_vae | VAE | Used only to encode image conditioning. T2VA does not use it; V3 never decodes with it. | |
| sampler | SAMPLER | — | |
| sigmas | SIGMAS | — | |
| sequence_prompt | STRING | Connect one Text (Multiline) for the complete sequence. | |
| prompt_mode | COMBO | Auto | 4 options: Auto, Fixed, List, Timeline |
| chunks | INT | 31–16 | — |
| chunk_seconds | FLOAT | 5.04–15 | — |
| width | INT | 134432–16384 | — |
| height | INT | 76832–16384 | — |
| continuity | COMBO | Balanced — 22 frames | 4 options: Auto — conservative, Balanced — 22 frames, Fast — 5 frames, Strong — 39 frames (Experimental) |
| base_seed | INT | 00–18446744073709550000 | — |
| first_frameopt | IMAGE | Optional. Leave image inputs disconnected for T2VA. | |
| prompt_overridesopt | H3_CONTINUUM_CLIP_OVERRIDES | — | |
| advancedopt | H3_CONTINUUM_ADVANCED_V3 | — |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| video_latents | LATENT | — |
| audio_latents | LATENT | — |
| assembly_plan | H3_CONTINUUM_ASSEMBLY_PLAN | — |
| result | H3_CONTINUUM_RESULT | — |