Mpromp2 — the AI Prompt GeneratorPrompt Library / Prompts for CogVideoX
Free Prompts for CogVideoX
This page collects every prompt in the library in its CogVideoX version — each one formatted the way CogVideoX takes prompts best, on its own page, all free. CogVideoX (Zhipu AI / THUDM) is an open-weights video model whose text encoder was trained on dense captions: it wants long, descriptive, caption-style sentences — a full paragraph describing the scene, the subject's appearance, the action and the camera, not a tag list. It does text-to-video and image-to-video at 720p, with no audio. The 2B checkpoint is openly licensed (Apache-2.0); the 5B carries the Tsinghua community license. Runs locally in Diffusers or ComfyUI.
Video
- Product Ad Video Prompt Prompt for CogVideoX — A short product video prompt with camera movement and pacing spelled out.
- Nature Scene Video Prompt Prompt for CogVideoX — A calm, realistic nature clip - the most reused stock-style video prompt there is.
- Character Animation Prompt Prompt for CogVideoX — A character doing one clear action - the reliable way to get usable AI video.
- Social Short (Vertical) Video Prompt Prompt for CogVideoX — A vertical, hook-first clip formatted for Shorts, Reels, and TikTok.
Image & Video
Logo-Safe Product Video (Clean Frame for Post) for CogVideoXA product video shot deliberately clean — empty center frame and a still final hold — so logo and headline can be added in post without fighting the footage.
UGC-Style Video Ad for CogVideoXA phone-shot, creator-style ad clip: handheld, everyday setting, one genuine reaction — made to look native to the feed, not produced.
B-Roll Footage Generator for CogVideoXClean, narration-ready B-roll: one subject, one gliding camera move, nothing that fights a voiceover or an edit.
Talking-Head / Avatar Setup Shot for CogVideoXA locked, centered presenter frame built for avatar and lip-sync workflows: even light, clean sides for captions, no drift.
Before / After Transformation Video for CogVideoXOne locked-off shot that carries a subject from its before state to its after state — the change is the whole clip.
Social Hook Clip (First-Frame Stopping Power) for CogVideoXA vertical clip engineered around the first frame: already mid-action at frame one, payoff held at the end — built to stop a scrolling thumb.
End-Card Shot (Closing Frame) for CogVideoXThe closing shot of a video or ad: subject centered, one light sweep, then a perfectly still hold with clean space for logo and call to action.- Logo-Safe Product Video (Clean Frame for Post) Prompt for CogVideoX — A product video shot deliberately clean — empty center frame and a still final hold — so logo and headline can be added in post without fighting the footage.
- UGC-Style Video Ad Prompt for CogVideoX — A phone-shot, creator-style ad clip: handheld, everyday setting, one genuine reaction — made to look native to the feed, not produced.
- B-Roll Footage Generator Prompt for CogVideoX — Clean, narration-ready B-roll: one subject, one gliding camera move, nothing that fights a voiceover or an edit.
- Talking-Head / Avatar Setup Shot Prompt for CogVideoX — A locked, centered presenter frame built for avatar and lip-sync workflows: even light, clean sides for captions, no drift.
- Before / After Transformation Video Prompt for CogVideoX — One locked-off shot that carries a subject from its before state to its after state — the change is the whole clip.
- Social Hook Clip (First-Frame Stopping Power) Prompt for CogVideoX — A vertical clip engineered around the first frame: already mid-action at frame one, payoff held at the end — built to stop a scrolling thumb.
- End-Card Shot (Closing Frame) Prompt for CogVideoX — The closing shot of a video or ad: subject centered, one light sweep, then a perfectly still hold with clean space for logo and call to action.
What this route costs — and how to pay less
Cost structure: open weights — CogVideoX runs on your own hardware, so the marginal cost per generation is zero. The same prompt is pre-configured in this library for every major model, so you can run it on whichever route fits your budget — comparing the structures before you generate is a free way to save money.
The layer above prompts
Prompts are single instructions. Skills are procedures your AI keeps — see the Skills Library for SKILL.md files and the setup guide for this platform.
Mpromp2 — the AI Prompt Generator. All prompts are free to use, including commercially. Full library: all prompts, all models.