FRAME

Console password

IMG

FRAME · Image Console

Free + paid image models behind one prompt-engineering layer
Images
Today
Free
Credits spent
Spend today

Brief

Empty means the image is generated with no writing at all.

Reference — who or what is in the picture

Camera & craft

Model

AI rewrite
Free fallback

Final prompt — editable, sent verbatim

The photograph

Nothing chosen yet.
Edit only the marked area
Everything outside stays exactly as it is.
Choose a photo to place the mask.
Safest — eyes and nose never touched
Drag the oval to position it. Keep it small. Anything inside is redrawn from scratch, so covering the eyes and nose is what makes a result stop looking like the person.

Shape & skin — instant, free, no AI

These move the photo's own pixels, so the likeness cannot change. Uses the oval above to know where the face is. Applied before anything is sent to a model.
Body shaping
Drag the blue box over the body, then use the sliders.
Blemish remover
Turn on, then click any spot on the photo above.

AI changes — makeup, hair, anything new

Only for things that are not in the photo. Slimming and skin are handled above, better and for free — don't ask the AI for those.

Before and after

Choose a photo to begin.
Missing cover only

Rows — one per line

Only the first column is required. Pipes, not commas — track titles contain commas. An artist whose name matches your Subjects roster automatically gets their reference photos.
Use {artist} {title} {genre} {mood}. Leave empty to build the scene from the columns.

Settings for every row

Style pack, model, resolution, ratio and detail are taken from the Generate tab, so set them up there first and check one image before committing a long run.
No batch running.

Add a subject

Two or three clear, well-lit photos of the same face from slightly different angles work far better than one. Avoid sunglasses, heavy shadow and group shots — the model has to be able to tell which face is the subject.

Why a roster

Saving someone once means every cover you ever make for them is built from the same reference bytes and the same identity wording. That is what keeps a face consistent across a whole catalogue — re-uploading photos ad hoc and re-typing the description each time is exactly what makes covers drift apart.

Reference photos work on the GPT Image models (billed per image by OpenAI) and on Nano Banana (free, but that Google account currently reports depleted image credits). The other providers need the photo at a public URL, which this console cannot serve from your PC.

Catalog

ModelProviderCostRatioPhotosStatusNotes

Providers

Keys live in .env next to package.json and never reach the browser — the console only learns whether each key is present. Copy .env.example to .env, fill in what you have, and restart the server.

How the layer works

1 · Style pack. Picks the photographer persona, the art direction of that genre, and the camera/lighting/colour defaults a professional would reach for.

2 · Dimensions. Every dropdown left on Auto defers to the pack. Override one and it wins — so you can nudge a single thing without rewriting the brief.

3 · AI rewrite (optional). Sends the composed prompt through Gemini wearing a senior-photographer hat, which fills in specifics a template cannot invent. Keeps every decision you made.

4 · Negatives. None of these four providers has a negative-prompt field, so the pack's avoid-list is folded into the prompt as one short trailing clause.

5 · Fallback. If a paid generation fails, the free providers are tried in turn. The reverse never happens — a free model failing will not silently spend credits.