How Midjourney reads prompts
Midjourney responds best to short, concrete phrases rather than long sentences. PromptLens analyzes your image and rewrites what it sees into that style, putting the subject first and following with style, setting, lighting, color and mood.
It also measures your image and adds the closest --ar value, so the composition of your result matches your reference. For photographic and product shots it adds --style raw to reduce Midjourney's default stylization.
Tips
- Remove phrases for details you don't want carried over before you paste the prompt.
- Use --no to exclude things, e.g. --no text, watermark.
- Change one phrase at a time when iterating, so you can see what each change does.
lone lighthouse on a sea cliff, storm clouds, crashing waves, cold blue light, dramatic wide shot, moody atmosphere --ar 16:9