What you can do with Mini Max H3 Director Cut Studio

4 use cases

• Plan a multi-shot AI video on a timeline

  Lay out Shots, Dialogue, Markers and Prompts on a multi-track timeline, then submit the valid media to ComfyUI.

  • Mini Max H3 Director Cut Studio
  • Content
  • Advanced
  View prompt

  Using the repo karuvanan/MiniMax-H3-Director-Cut-Studio, help me place the five H3 model files for video_minimax_h3_r2v_9image_3audio_3video_api.json into the ComfyUI folders named in the README, then build a three-shot timeline for <project name>. Expected output: a list of the generated shot prompts. Verify that each model file name matches the README exactly before submitting to ComfyUI.

  Share Save

• Swap a cast member in an existing source video

  Replace a person in a source video shot by shot using the Character-Swap LoRA, the H3 Relay and SAM3.1 masks.

  • Mini Max H3 Director Cut Studio
  • Content
  • Advanced
  View prompt

  Using the repo karuvanan/MiniMax-H3-Director-Cut-Studio, set up the Source Video Cast Remake H3 path for the source clip <source video path> and swap the role named <role name> with the reference image <P picture path>. Expected output: a regenerated shot in the studio with the swap applied to that role only. Verify that LabsH3PrepareVideo, H3RelayWindowedEdit, SAM3_VideoTrack and SAM3_TrackToMask appear in ComfyUI without IMPORT FAILED.

  Share Save

• Generate Design concept images with Z-Image and LM Studio

  Create concept reference images with the Z-Image workflow and write prompts through a local LM Studio model.

  • Mini Max H3 Director Cut Studio
  • Content
  • Intermediate
  View prompt

  Using the repo karuvanan/MiniMax-H3-Director-Cut-Studio, configure the Design page to use the LM Studio Base URL http://127.0.0.1:1234/v1 and a Model ID that I select from the live list, then generate one 1024 by 576 concept image for <scene description>. Expected output: one Z-Image concept image and its caption. Verify that TEST CONNECTION succeeds before generating.

  Share Save

• Split a song into vocal and music tracks locally

  Run the Kim_Vocal_2 audio separator on a CUDA GPU to output Vocal, Music and original Mix files.

  • Mini Max H3 Director Cut Studio
  • Content
  • Intermediate
  View prompt

  Using the repo karuvanan/MiniMax-H3-Director-Cut-Studio, install the CUDA runtime and start the Audio Separator server, then split <audio file path> into Vocal and Music files. Expected output: a Vocal file, a Music file and the original Mix. Verify that the Vocal and Music durations and channel counts match the input before use.

  Share Save

How to use Mini Max H3 Director Cut Studio

1. Check you have what it needs: A Windows PC with ComfyUI installed, plus a high-VRAM machine if you want to run the local server roles described in the README; Completed Step 1 deployment (Studio Runtime, Qwen3-TTS and VoxCPM2) from the README's step1.md; The model files listed in the README, saved under the exact file names it gives.
2. Open karuvanan/MiniMax-H3-Director-Cut-Studio on GitHub and read the README and license.
3. Complete Step 1 in the README (step1.md) on the machine that will run the studio.
4. Download the model files for video_minimax_h3_r2v_9image_3audio_3video_api.json (UNETLoader, CLIPLoader, two VAELoaders and the H3 Turbo LoRA) and place them in the ComfyUI folders the README lists.
5. Start the studio from the project root with .\run_h3_prompt_studio.bat.
6. Add images, video and audio to the Virtual Media Pool, then build Shot, Dialogue, Marker and Ending Hold items on the Timeline.

These steps come from our top use case, Plan a multi-shot AI video on a timeline, which also has a copy-paste agent prompt.

Mini Max H3 Director Cut Studio alternatives · Best AI repos for Content

About this repo

• Who it is forFilmmakers and creators with a Windows PC, ComfyUI and model files who want a timeline-based workflow for AI video.
• Why it is hot
• LanguagePython
• Sourcegithub.com/karuvanan/MiniMax-H3-Director-Cut-Studio · Back to Hot repos

Requirements

• A Windows PC with ComfyUI installed, plus a high-VRAM machine if you want to run the local server roles described in the README
• Completed Step 1 deployment (Studio Runtime, Qwen3-TTS and VoxCPM2) from the README's step1.md
• The model files listed in the README, saved under the exact file names it gives
• Model and source-footage licenses checked by you; the README says weights are not included in the source code

Related repos

• Trended earlier Hermes Speech Hermes Speech is a Python plugin that gives the Hermes Agent native text-to-speech and speech-to-text through AllModels, with voice search, previews, and conversational setup.
• Trended earlier Wonder Pill Wonder Pill is a Claude skill that turns a topic into an interactive mind map of inverted assumptions, with gut-check notes and no ranked answers.
• Trended earlier Comfy UI Mini Max H3 Image Studio A ComfyUI custom node pack with two starter workflows for MiniMax H3 text-to-image generation and image editing, with optional stackable LoRAs.