Nodes/noEmbryo nodes/H3 Motion Context Clip Stitcher /noEmbryo
ComfyUI Node

H3 Motion Context Clip Stitcher /noEmbryo

Final assembly for NikoDemon80's H3 Motion Context AV clip archives. Loads numbered h3_motion_context_av_v1 files, decodes one clip at a time, dissolves the overlap between adjacent clips (video + synchronized audio), and concatenates them to a final video and audio stream.

By noembryo·Created 3 years ago·Updated about 14 hours ago· 29
H3 Motion Context Clip Stitcher /noEmbryo
  • video_vae
  • audio_vae
  • latent
  • images
  • audio
  • frame_count
  • report
folderh3_context
patternclip_*.safetensors
first_clip1
last_clip0
context_length22
fps24.000
Categoryvideo/minimax

Inputs (9)

NameTypeDefaultDescription
folderSTRINGh3_contextFolder containing clip_00001.safetensors, clip_00002.safetensors, etc. Absolute paths and paths relative to ComfyUI/output are accepted.
patternSTRINGclip_*.safetensorsFilename glob. The final five-digit number is treated as the clip index.
first_clipINT11–9999First approved clip to include.
last_clipINT00–9999Last clip to include. 0 = every clip from first_clip onward.
context_lengthCOMBO22Number of decoded frames to crossfade at each clip boundary. The normal setting is 22 frames. This is the overlap length that is dissolved between adjacent clips. 5, 22, 39 or 56 are the lengths that are a whole number of latent steps, which is why other numbers aren't offered.
fpsFLOAT24.0001–240H3 native output rate. Keep this at 24 unless your workflow deliberately changes it.
video_vaeoptVAEMiniMax H3 video VAE (FP16 or INT8 ConvRot).
audio_vaeoptVAEMiniMax H3 audio VAE FP32. Required for the AUDIO output.
latentoptLATENTOptional: the currently-generated AV latent (from your H3 sampler), used in place of the highest-numbered file on disk.

Outputs (4)

NameTypeDescription
imagesIMAGE
audioAUDIO
frame_countINT
reportSTRING