Lumina Image Studio: a prompt-to-image generation interface

Lumina is an interface concept for generating illustration assets from text prompts and keeping them in a searchable library. The Create view puts a prompt bar, a model picker (the mock list names Nano Banana, GPT Image and Kling image models), aspect-ratio and batch-size controls, a private-mode toggle, a reference-image drop zone and a style library (ultra-flat, single-line, sketch, impasto oil, Japanese anime and more, each with example prompts) on one screen. Pressing generate runs a simulated progress bar and adds a generation card to the history; the Library view lists past generations under tabs with a search box.

The problem and who it is for

The problem it explores: an explainer-video pipeline needs hundreds of on-style assets, and the people making them switch between models and styles constantly. Keeping model, style and prompt together avoids bouncing between three tools. It is for the artist or producer generating assets, not the model provider.

Stack and status

Stack observed in the source: React 18 with in-browser JSX and inline styles, no build step, no backend and no real model calls; generation is a timer and the results are placeholder graphics. Status: a UI prototype. It sits alongside the other explainer-video tools on the homepage.

Live demo

Prototype status: UI demo · models, styles, inspirations, library items and every generated image are mock data; nothing is sent to a model

Type a prompt and press generate to see the simulated flow, or open Library from the icon sidebar. The demo needs JavaScript and a desktop-width window.