Read all about the journey to make this HERE

MY LC123 Pack, AV Pack, Vision pack, and maskmaker pack all must be up to date for this to work. You may have to get them from GitHub directly

What is it?

• A multi-scene workflow for MiniMax H3. You write the whole movie on one node: who is in it, where it happens, every shot.

• It makes the shots one after another and joins them into one movie, with the sound and the dialogue.

• Cast, outfits, places, props, storyboard frames and post processing all live in one workflow. No copying clips between tabs, no video editor.

Why is it better?

• No seams. 🎞 Keep rolling carries the camera straight on from the shot before. Watch shots 1 to 2 and 5 to 6 in Coffee Date and try to find the cut.

• No sound issues. The sound carries across the join, every line lands at its own second, and a silent shot stays silent. No babble, no voice from nowhere.

• The same faces, start to finish. Headshots, outfit pictures and a memory of the first shot keep everyone looking like themselves.

• Redo one take. Don't like shot 4? ↻ makes that one again. Edit a shot and only that shot is made again on the next Run. The rest stays.

• Rooms that stay rooms. A 🌐 360 place turns the camera around inside the same room, so a close-up across the table shows the room behind them, not a new one H3 invents.

• Openings and endings you choose. Pin a 🖼 start frame or a 🏁 end frame and the shot starts or lands exactly there.

• Outfits that hold. Change clothes on any shot and they stay changed.

• A LoRA for one shot. 🎛 Put a character, outfit or style LoRA on one shot only.

• It tells you before it breaks. Every shot shows the exact text H3 gets, with warnings before you spend time rendering.

How does it work?

The full walk-through is in the guides:

• 📘 User Guide: https://github.com/lonecatone23/ComfyUI_LC_AV_nodes/blob/main/docs/showrunner/USER_GUIDE.md

• ✍️ Prompting Guide: https://github.com/lonecatone23/ComfyUI_LC_AV_nodes/blob/main/docs/showrunner/PROMPTING_GUIDE.md

The short version:

The workflow

🎭 Cast: a headshot and an outfit picture for each person. Make them with any image model, or pick from the free 📚 Library.

📍 Places: a picture of each room, or a 360 panorama.

🎞 Shots: one card per shot, top to bottom is the movie. The board on top plays, remakes and locks each one.

🎥 Make: one Run makes every shot that isn't made yet and joins the movie.

🎞️ Post: frame interpolation to 48 fps, RTX deblur and upscale, sound.

💡 It opens with a sample story. Hit Run before you change anything. Then take the guided tour (Vision Assistant → Tour).

Size Requirements:

Instagram: https://www.instagram.com/synth.studio.models/

Buy me a ☕ https://ko-fi.com/lonecatone

This represents hundreds of hours of work. If you enjoy it, please 👍 like, 💬 comment, and feel free to ⚡ tip 😉

"True Nothing is. Permitted Everything is"- Yoda Auditore, Assassin's Wars