Skip to main content
Short-form explainers (YouTube Shorts, Reels, TikTok) use the same compose_video pipeline as long explainers, with tighter scene counts and optional 9:16 output. Defaults that work well for shorts: 4–5 scenes, 30–90s total, --no-text (voiceover only), hybrid B-roll or --ai-images-only.
See also: Compose video for hybrid mode, custom script APIs, and env vars.

Quick recipe (LLM script)

Hybrid mode (default) mixes stock video and photo B-roll per scene. Add --ai-images-only when you want Gemini/Pollinations images instead of stock photos (requires GEMINI_API_KEY or POLLINATIONS_API_KEY).

Terminal tool short (custom script + CTA)

For repeatable shorts (intro → value → install tip → like/subscribe), use a JSON script instead of --llm. Keep narration punchy; last scene is your CTA. terminal-short-script.json:
Set media_type per scene when you want explicit control (video for motion B-roll, image for diagrams or UI stills). See Compose video.

Vertical 9:16 (Shorts / Reels)

compose_video has no --orientation flag yet. Set env vars for portrait output and vertical stock search:
For full AI video clips (not stock B-roll), use ai_video with VIDEO_ASPECT=9:16:

Useful flags & env

Natural language also routes correctly: