Image → Prompt
Drop a picture. A vision model reads it field by field — subject, wardrobe, setting, light, lens, palette, composition, mood — and writes the prompt that rebuilds it. Edit any field; the prompt rewrites itself.
Sign in to the Atelier
Free, forever. An account keeps your reading history and your saved previews on this device.
No account yet?
Accounts live in this browser only — there is no studio server holding your password. It keeps your history private to this device, but it is not bank-grade security: don't reuse an important password.
Drop a picture
or tap to choose · JPG, PNG, WebP, GIF · up to 8 MB
Used in the prompt and in the generated preview.
The reading appears here.
Generate an image from the prompt above, in your chosen aspect ratio. Edit a field, generate again, compare.