Camera Language and the Prompt Grammar
- →Distinguish the two officially documented prompt layers - casual bracket hints and the trained natural-language camera vocabulary - and know which official source documents each
- →Compose camera direction as motion type + amplitude + speed sentences from the 12-type vocabulary
- →Read and hand-author the structured prompt format: shot markers, timestamped cuts, speaker IDs, dialogue tags, and continuity tags
H3 has no camera parameter. Every camera decision you make rides inside the prompt — and MiniMax documents two distinct systems for expressing it. Nearly every guide in circulation teaches a mangled version of the first and doesn't know the second exists.
Layer one is the casual syntax on the platform's video-generation guide: bracket tags placed directly after key descriptions. Verbatim: "add camera motion instructions (e.g., [pan], [zoom], [static]) directly after key descriptions to guide the camera work." Note the "e.g." — the docs present [pan], [zoom], [static] as examples, not a closed vocabulary.
Unlock “Camera Language and the Prompt Grammar”
The 4-lesson Foundation module is free and always will be — you have already read the part most guides get wrong. The paid lessons go deeper, with labs, knowledge checks, and every technical claim verified against live documentation.
Already bought it? Sign in with the email you used at checkout.