Directing note: saying what not to show
Four models document a negative prompt field for video; one maker advises against negative prompts altogether. As of 2026-09-25.
The reasoning behind the advice is simple: a model that reads a word may draw it, even after a no.
Where no field exists, describing what the frame does contain is the safer route. A frame described in full, subject, setting, light and camera, leaves the model less room to add what was not wanted.
When a stray element keeps appearing, a first or last frame that shows the scene without it is another way to steer the model away from it.
1From the docs
Negative prompts are a tool that helps specify the elements that you don't want generated in your video.
Google Cloud, Veo video prompt guide
Negative prompt for video generation
Tencent Hunyuan, HunyuanVideo-1.5 repository
Negative prompts can be inputted for Text to Video by enabling the Negative Prompt toggle found in Advanced Settings.
Leonardo.Ai Help Center, generating video
- Negative promptsNegative prompts list elements that should not appear in the video
- Command line argumentsThe inference script takes a negative prompt, empty by default
- Negative PromptingA negative prompt, switched on under Advanced Settings; the help advises writing what to avoid without no or without
- TipsThe guide advises positive phrasing and avoiding negative prompts
- Use a negative promptA negative_prompt field excludes elements from the video
- FrameA first and a last keyframe image can be uploaded as fixed points for the clip
- Camera angles and shot typesShot sizes and angles are defined one by one, from extreme close-up to wide, low-angle and dutch angle
- AbstractText-to-video and image-to-video in one framework
2A shot list from a script
Where “Saying what not to show” has to be written by hand, SceneMixer's pipeline writes the shots; its guides explain:
SceneMixer turns a novel into a watchable short-drama series in one pipeline: upload the novel (Chinese or English) → AI extracts characters, scenes, and an episode-by-episode structure → generates consistent character and scene art → writes storyboard scripts → renders video segment by segment (16:9 by default, 9:16 in one setting) → composites full episodes.
SceneMixer, Turn a Novel into a Video Series with AI (read 2026-10-13)
SceneMixer writes dialogue into the script analysis, assigns each line to a shot with a speaker and a short delivery note, and the video model performs it — picture and audio generated together, so the mouth forms that sentence in the language you chose.
SceneMixer, AI Dialogue Generator (read 2026-10-13)
3Other notes
Examples are quoted from the models' docs; no model was run.