Prompts for Images Versus Prompts for Text
Learn why image generators need a different kind of prompt from chat assistants, and how to write both.
A taste of a lesson
I asked for 'a beach with no people' and got people in the background every time. Why?
Many image models do not handle 'no' well. They pick up the word 'people' and it nudges the picture towards including them. Describe what you want to see instead: 'an empty, deserted beach at sunrise, smooth untouched sand, calm sea, wide shot'. Words like empty and deserted carry the meaning without naming people. If your tool has a separate negative prompt box, put 'people' there instead. Try rewriting your prompt using only things that are in the scene. What time of day do you want the beach to feel like?
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Explain why image prompts describe a picture while text prompts give instructions
- Write an image prompt covering subject, composition, lighting and style
- Work around weak spots like negation, counting and text inside images
- Iterate on an image by changing one element at a time
Lesson plan
- 1 Instructions versus descriptions See why text models want instructions and image models want a picture described. Start
- 2 The parts of an image prompt Build prompts from subject, setting, composition, lighting and style. Start
- 3 Known weak spots Avoid common surprises with negation, counting and text. Start
- 4 Iterating with control Improve an image step by step without losing what worked. Start
- 5 Ethics and permissions for images Know the questions to ask before using generated images. Start
Try asking
About this tutor
People who write good prompts for chat assistants are often surprised when the same habits fail with image generators. Text models follow instructions and conversation; image models mostly respond to description: subject, setting, composition, lighting, style and format. This tutor teaches the differences side by side. You practise turning an instruction into a visual description, learn why 'no cars' can produce cars, why text inside images is unreliable, and how to change one element at a time when iterating. You also cover the ethics that differ for images, such as real people's likenesses and named artists' styles. Works with any image tool.
Reviews
Students can review a tutor after a paid lesson. Nobody has yet.
About the teacher
Teaches prompting one technique at a time, with plenty of before and after examples
9 tutors 339 lessons taught Sample
I teach people to write prompts the way I once taught people to write briefs: say what you want, who it is for, and what good looks like. I came to AI from editing and corporate training, so I care more about clear thinking than clever tricks. In my lessons we take one technique at a time, try it on...
See Chiara's profile and tutorsMore like this
Other tutors on the same or nearby topics.