Nodes/ComfyUI-NynxzH3/H3 Latent Slice
ComfyUI Node

H3 Latent Slice

Slices an H3 AV latent along the latent-frame axis and drops or trims the audio stream to match. For a single frame as an image use MiniMax H3 Decode Frames — a lone latent token does not decode to the frame it was inside a clip.

By Nynxz·Created 14 days ago·Updated 5 days ago· 4
H3 Latent Slice
  • samples
  • LATENT
  • latent_frames
start0
count0
audiodrop
CategoryNynxz/H3/Latent

Inputs (4)

NameTypeDefaultDescription
samplesLATENTAn H3 latent, nested (video+audio) or bare.
startINT0-4096–4096First latent frame to keep. Negative counts from the end.
countINT00–4096How many latent frames to keep. 0 means everything from start on.
audioCOMBOdrop'drop' returns video only, which is what a VAE Decode wants. 'trim' keeps a proportionally cut audio stream so the result stays a playable AV latent.

Outputs (2)

NameTypeDescription
LATENTLATENT
latent_framesINTLatent frames in the result.