Nodes/H3 Continuum/H3 Continuum Sampler V3.5
ComfyUI Node

H3 Continuum Sampler V3.5

H3 Continuum V3.5 sampler. It preserves the V3.4 outputs and adds one physical-group refine context for context-aware Hi-Res Fix or an external latent Second Pass.

By ukr8b3g-cmyk·Created 15 days ago·Updated about 8 hours ago· 21
H3 Continuum Sampler V3.5
  • model
  • clip
  • video_vae
  • sampler
  • sigmas
  • first_frame
  • last_frame
  • reference_image_1
  • reference_image_2
  • reference_image_3
  • reference_video_1
  • driving_audio
  • audio_vae
  • reference_audio_1
  • reference_audio_vae
  • video_latents
  • audio_latents
  • assembly_plan
  • status
  • driving_audio
  • refine_context
sequence_prompt
prompt_modeAuto
chunks3
chunk_seconds5.0
width1344
height768
continuityBalanced — 22 frames
base_seed0
audio_continuitytrue
diagnosticsBasic
reroll_from_chunkAuto
reroll_nonce0
strict_compatibilitytrue
debugfalse
show_previewtrue
run_storageOff
run_name
reference_sizeMatch Output
project_id
video_reference_sizeEfficient - 0.4 MP
CategoryMiniMax H3/Continuum

Inputs (35)

NameTypeDefaultDescription
modelMODEL
clipCLIP
video_vaeVAEUsed only to encode image conditioning. T2VA does not use it; Continuum never decodes with it.
samplerSAMPLER
sigmasSIGMAS
sequence_promptSTRINGConnect one Text (Multiline) for the complete sequence.
prompt_modeCOMBOAutoAuto accepts Fixed, list-separated, and timeline prompt styles.
chunksINT31–16Number of sequential Continuum chunks to generate.
chunk_secondsFLOAT5.04–15Target duration per chunk. Five seconds is the validated default.
widthINT134432–16384Output width. Use a multiple of 32.
heightINT76832–16384Output height. Use a multiple of 32.
continuityCOMBOBalanced — 22 framesAmount of prior video context retained at each chunk boundary.
base_seedINT00–18446744073709550000Base seed used to derive deterministic per-chunk seeds.
audio_continuityBOOLEANtrueOn passes prior audio context into continuation chunks. Turn it off only to isolate or replace generated audio.
diagnosticsCOMBOBasic3 options: Basic, Detailed Report, Off
reroll_from_chunkCOMBOAutoAuto resumes the longest compatible saved prefix. Choosing a chunk reuses earlier chunks and regenerates that chunk and everything after it.
reroll_nonceINT00–4294967295Change only when regenerating an explicit chunk and you want a new variation with otherwise identical settings.
strict_compatibilityBOOLEANtrue
debugBOOLEANfalse
show_previewBOOLEANtrue
run_storageCOMBOOffAtomically save raw AV chunks and resume a compatible saved run.
run_nameSTRINGEnter a stable name for this saved run. Compatible chunks are selected automatically.
reference_sizeCOMBOMatch OutputMatch Output is the practical default; Max Identity preserves more reference detail.
project_idSTRINGOptional. Leave blank to derive a stable ID from this sampler node. Run Name remains the explicit override.
video_reference_sizeCOMBOEfficient - 0.4 MPEfficient limits Video Guide Frames to about 0.4 MP; Balanced uses about 0.6 MP; Match Output uses the output pixel area. Source aspect ratio is preserved and smaller sources are not enlarged.
first_frameoptIMAGEOptional. Leave all image inputs disconnected for T2VA.
last_frameoptIMAGE
reference_image_1optIMAGE
reference_image_2optIMAGE
reference_image_3optIMAGE
reference_video_1optIMAGEOptional video guide. Connect the IMAGE frame batch from a video loader; source-video audio is not included. Frames are interpreted at 24 fps and applied to every chunk. Non-native frame counts are padded by repeating the final frame to the next H3 17k+5 count, up to the one-chunk limit.
driving_audiooptAUDIOOptional original audio timeline. It is used as native H3 guide conditioning and selected unchanged for final output.
audio_vaeoptVAERequired only when Driving Audio is connected. Uses the same Audio VAE encode path as ComfyUI Core MiniMax H3 Add Guide.
reference_audio_1optAUDIOOptional standalone audio reference for H3 conditioning. It is not the audio track of Video Guide Frames. Unlike Driving Audio, it does not replace the generated final audio.
reference_audio_vaeoptVAERequired only when Reference Audio is connected. It encodes the reference for H3 conditioning; generated audio remains the output.

Outputs (6)

NameTypeDescription
video_latentsLATENT
audio_latentsLATENT
assembly_planH3_CONTINUUM_ASSEMBLY_PLAN
statusSTRING
driving_audioAUDIO
refine_contextH3_CONTINUUM_REFINE_CONTEXT