All categories/Image and video
Text to image · Level Starter · Image Variants Workflow
Variants: ten options at once
The client wants several directions and one output per run is too slow.
Prompt
You describe only what can be seen: subject, action, placement, light, lens, material, aspect. You never write 'premium feel' — nothing there can be drawn.
Task: The client wants several directions and one output per run is too slow.
[Before you start, do three things]
1. Restate the goal in one sentence: who reads the output, and what decision it serves.
2. If I have not given you the following, ask me. Do not assume:
- what must not change
- what may change
- aspect and count
- client preference
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.
[Go through each of these]
- fix the constants first
- vary one dimension per run, with a list of dimensions
- a numbering scheme so the client can point
- say which variants are worth producing
[Process]
1. Confirm use and aspect first, then subject and composition
2. Describe placement, relative size and depth order, not adjectives
3. Light: direction, hardness, temperature. Lens: focal length and depth of field
4. Negative list separate
5. Two variant directions for comparison
[Output]
A positive description, a negative list, and parameter suggestions: aspect, style, texture.
[Do not]
- No abstract words that cannot be drawn
- Never three subjects in one frame
- Do not stack style words for quality
- Never ignore aspect ratio versus intended useWhat to state before it starts
Leave these unsaid and the model fills them with an assumption. Most 'wrong answers' start here.
- what must not change
- what may change
- aspect and count
- client preference
Go through each of these
Where this kind of work most often goes wrong. Already in the prompt, so you do not have to remember it.
- fix the constants first
- vary one dimension per run, with a list of dimensions
- a numbering scheme so the client can point
- say which variants are worth producing