wiki / images-midjourney-campaign-style-reference / edit

Suggest an edit.

Improvements are reviewed before going live, like any wiki.

Title

One-line summary

The prompt

This entry is Midjourney-specific: the double-dash parameters below are Midjourney syntax and will not work as written in DALL-E, FLUX, or Stable Diffusion.

Step 1 — generate the anchor image:
[HERO SUBJECT AND ACTION, E.G. A CYCLIST CRESTING A HILL], [ENVIRONMENT, E.G. COASTAL ROAD AT DAWN], [LIGHTING, E.G. LOW SIDE LIGHT WITH LONG SHADOWS], shot on [CAMERA OR FILM STOCK, E.G. KODAK PORTRA 400], muted [PALETTE DESCRIPTION, E.G. TEAL AND SAND] palette --ar 4:5 --style raw --no text, watermark, logo, border

Upscale the strongest result, copy its image URL, and note the seed reported for that job.

Step 2 — every further image in the set:
[NEW SUBJECT OR SCENE FOR THIS IMAGE], [SAME ENVIRONMENT FAMILY AS THE ANCHOR], [IDENTICAL LIGHTING WORDING FROM STEP 1], shot on [SAME CAMERA OR FILM STOCK], muted [SAME PALETTE DESCRIPTION] palette --ar 4:5 --style raw --sref [ANCHOR IMAGE URL] --sw 200 --seed [ANCHOR SEED NUMBER] --no text, watermark, logo, border

Keep the wording of the lighting, camera, and palette clauses identical across every image; change only the subject clause. Raise --sw toward 400 if the set drifts apart visually, or lower it toward 100 if new images copy the anchor's content instead of just its look. Keep --ar fixed for the whole campaign so crops match across placements.

Write it so someone can paste it straight into Claude, ChatGPT or Gemini and fill the [BRACKETS].