MiniMax H3 Prompt Template: Multi-Character Video Generation
ZerOne82 · reddit · 2026-08-22
The author shares a prompt template for MiniMax H3 that generates multi-character video dialogues using a specific structure.
The template consists of three parts:
- definitions: Maps character codes (e.g., <S1>) to visual descriptions and dialogue lines.
- scene: Defines the environment (lighting, background, camera angle).
- shot: Sequences the shots, specifying which character speaks which line.
A demo featuring Brad Pitt, Angelina Jolie, and Mr. Bean is provided. Recommended workflow: Model (FL2V) -> LoRA(4s-Lightx2v SLA) -> Comfy attn -> Shift(12,3) -> KSampler(6 steps, euler+simple). Note: Avoid SLA, SLA2, or cache as they interfere with the output.
More from Multimodal
- Inside 'The Very Next Day': 2D keyframes plus a custom IC-Lora by LTX — paultrillo · 2026-10-16
- fal Worlds: designer unveils brand identity for fal's generative world-models platform — adamho · 2026-10-16
- Slopcannon picks Muse Wonder as default AI director, adds frontier model picker — basedjensen · 2026-10-16
- Developer mass-produces 200 videos in hours using Claude Code + Hyperframes — mcraddock · 2026-10-16
- Creator showcases V2 of AI-generated artwork 'Alien Rebel' — creatoroff · 2026-10-16
- Mirage Tesseract lets Claude and ChatGPT create full HDR videos and motion graphics — LinusEkenstam · 2026-10-16