Nodes/ComfyUI-FL-MiniMaxH3/FL MiniMax H3 Beat Shot Planner
ComfyUI Node

FL MiniMax H3 Beat Shot Planner

Turns an FL beat prompt schedule into MiniMax H3 renders. Prompt sections grouped in the audio sequencer share one render; ungrouped sections remain independent.

By filliptm·Created 19 days ago·Updated a day ago· 0
FL MiniMax H3 Beat Shot Planner
  • clip
  • vae
  • audio_vae
  • prompt_schedule
  • timeline_audio
  • prompt_envelopes
  • ref_images
  • ref_videos
  • ref_video_audios
  • ref_audios
  • shot_plan
  • scheduled
  • latent
  • semantic
length124
global_prompt
width1344
height768
affect_audiovideo only
ref_image_sizematch
visual_condition_fidelity1.00
visual_reference_modefull
reference_strength1.00
CategoryFL/MiniMax H3/Prompting

Inputs (19)

NameTypeDefaultDescription
clipCLIPMiniMax H3 text encoder used for each planned render.
vaeVAEMiniMax H3 video VAE used to encode image and video references.
audio_vaeVAEMiniMax H3 audio VAE used to encode each shot's matching audio slice.
lengthINT1245–3600Requested frame count for a single render, used only when prompt_schedule is not connected. H3 snaps it to the 17k+5 frame grid.
global_promptSTRINGPersistent character, identity, visual style, and production context.
widthINT134432–16384Width of every planned render.
heightINT76832–16384Height of every planned render.
affect_audioCOMBOvideo onlyChoose whether scheduled prompt masks also affect H3 audio tokens.
ref_image_sizeCOMBOmatchReference image sizing, matching the standard MiniMax H3 reference node.
visual_condition_fidelityFLOAT1.000–1Noise fidelity for H3 visual latent references and motion context. Lower values replace more latent detail with seeded noise. This does not scale Qwen-VL.
visual_reference_modeCOMBOfullSelect which visual reference paths receive connected images and videos. Reference audio and video soundtracks remain active in every mode.
reference_strengthFLOAT1.000–1Blend from a prediction without connected visual references at 0 to the selected mode at 1. Audio and prior-shot motion context remain in both. Beat sampler cfg multiplies this value; effective blends other than 0 or 1 require both predictions while sampling.
prompt_scheduleoptFL_PROMPT_SCHEDULEExact 24 fps shot ranges from FL Audio Beat Prompt Schedule. Optional; when disconnected the node plans one full-length render from the length input.
timeline_audiooptAUDIOFrame-aligned cropped audio from FL Audio Beat Prompt Schedule. Each shot receives only its matching slice. Optional; when disconnected each planned render has no shot-local audio reference.
prompt_envelopesoptFL_PROMPT_ENVELOPE_SETSequence-wide reactive prompt envelope set from FL Audio Beat Prompt Sequencer, sliced and rebased for every planned render.
ref_imagesoptCOMFY_AUTOGROW_V3Character or scene images applied to every planned render.
ref_videosoptCOMFY_AUTOGROW_V3Reference videos applied to every planned render.
ref_video_audiosoptCOMFY_AUTOGROW_V3Soundtracks paired by index with reference videos.
ref_audiosoptCOMFY_AUTOGROW_V3Additional global audio references applied after the shot-local timeline audio.

Outputs (4)

NameTypeDescription
shot_planFL_H3_SHOT_PLANPlanned nested H3 latents and conditioning for sequential sampling.
scheduledCONDITIONINGPositive conditioning. Meaningful for the single render produced when prompt_schedule is not connected; with a schedule it reflects the first render.
latentLATENTNative nested MiniMax H3 video/audio latent. Meaningful for the single render produced when prompt_schedule is not connected; with a schedule it reflects the first render.
semanticCONDITIONINGSingle global conditioning. Used for the manual single render; with a schedule it reflects the first render.