Interactive Tool

P.A.S.S. Prompt Generator

Build structured transformation prompts, upload a photo, and generate the reimagined result with Gemini or OpenAI.

Scene & Source Image

Source Photograph (JPEG or PNG — up to ~8 MB)

Choose model Google · 1 credit Gemini 3 Pro Image (Nano Banana Pro) — highest fidelity Google model that accepts your input photo. OpenAI Images 2.0 · 1 credit GPT Image 2 — agentic model that plans before generating, with 4K support and the strongest instruction-following / text rendering. Google + OpenAI (Both) · 2 credits · returns 2 images Run both providers in parallel and compare. Failed side is auto-refunded.

Standard (1 credit) is either Google or OpenAI; Both runs them together (2 credits); Lane C can add FLUX (+2 credits).

Scene Type

Scene Description (brief description of what's in the photo)

Transformation Type

Primary Transformation

Divergence Lane

Lane A Purist

Plausible enhancement only. Near-documentary realism.

Lane B Storyteller

Visible but believable environmental change.

Lane C Visionary

Creative reimagination. Labeled as interpretive art.

P.A.S.S. Refinements

Optional — override auto-generated values

P Preserve

A Atmosphere

S Surfaces

S Subtleties

P.A.S.S.+ Controls

Extended prompt modifiers

Optics

Color Discipline

Post Character

Exclusions

Realism Target

Configure & Generate

Fill in the form on the left and upload a source photo. Sign in and click "Generate" to run the transformation.