ComfyUI Node
[Legacy] H3 Continuum Join
Prepare a native MiniMax H3 continuation clip from a previous AV latent state. Uses one timeline-positioned video/audio context reference, per-MODEL wrappers only, and composes with external SageAttention, Sol-Attn, and Spectrum.
[Legacy] H3 Continuum Join
- model
- conditioning
- latent
- previous_state
- model
- conditioning
- latent
- plan
- report
◄continuityBalanced — 22 frames►
◄extend_seconds5.0►
◄audio_continuitytrue►
◄first_frame_policyUse Continuum context; keep visual identity tokens (Recommended)►
◄preserve_last_frametrue►
◄strict_compatibilitytrue►
◄debugfalse►
CategoryMiniMax H3/Continuum/Legacy
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| conditioning | CONDITIONING | — | |
| latent | LATENT | — | |
| continuity | COMBO | Balanced — 22 frames | 3 options: Balanced — 22 frames, Fast — 5 frames, Strong — 39 frames (Experimental) |
| extend_seconds | FLOAT | 5.00.5–15 | Requested NEW duration. H3 snaps to the nearest 17k+5 frame count. |
| audio_continuity | BOOLEAN | true | Carry the previous native audio latent and place it on the new clip timeline. |
| first_frame_policy | COMBO | Use Continuum context; keep visual identity tokens (Recommended) | Recommended: replace only the old spatial keyframe. If the prompt was encoded with an input image, its Qwen visual tokens remain as an identity cue. |
| preserve_last_frame | BOOLEAN | true | Move an existing FL2VA last-frame anchor to the end of the continuation clip. |
| strict_compatibility | BOOLEAN | true | Run proactive native-H3 API checks. Marked Continuum payload failures always stop; they are never silently bypassed. |
| debug | BOOLEAN | false | — |
| previous_stateopt | H3_CONTINUUM_STATE | — |
Outputs (5)
| Name | Type | Description |
|---|---|---|
| model | MODEL | — |
| conditioning | CONDITIONING | — |
| latent | LATENT | — |
| plan | H3_CONTINUUM_PLAN | — |
| report | STRING | — |