AI Image Generator for Study Visuals

From a text prompt describing what you want — like "a labeled diagram of the human heart" — and, in the AI Image Playground, from up to 3 reference images you upload or pick from your content library. In-page and flashcard generation are text-prompt only; reference images are a Playground feature.

It depends where you generate. From inside a page or flashcard, the image you pick is attached directly to that study component. From the Playground, images are saved with their prompt, model, preset, and aspect ratio — downloadable on paid plans. In chat, the image is embedded right in the conversation.

In pages and flashcards you choose between three styles — Cartoon, Realistic, and Sketch. The AI Image Playground goes further: pick one or more models — GPT Image, Grok Imagine, and Nano Banana 2, alongside the default Nano Banana 2 Lite — plus study-ready presets (like labeled diagrams or anatomy) and an aspect ratio (1:1, 16:9, 9:16, 4:3, or 3:4), or skip the controls entirely and describe what you want in the Playground chat.

Generated images aren't editable directly. To change one, regenerate with a new prompt, or in the Playground use it as a reference image and remix from there. The Playground's reference-image workflow is the way to iterate on a visual you've already made.

Yes — it's free to try. Each new image or edit in the AI Image Playground uses one AI creation per selected model. Images created or edited in regular AI Chat do not use AI creations; the usual Chat message allowance still applies.

In the Playground, each selected model produces one image. Select several models to compare them side by side; each output uses one AI creation. Other image tools can return their own set of candidates.

Yes, in the AI Image Playground. Upload 1–3 reference images from your device, or select images you've made before from your content library, and generate edits or variations from them. In-page and flashcard generation are text-prompt only.

In-page and flashcard generation are the fast path: type a prompt, pick Cartoon, Realistic, or Sketch, and the image attaches to your study component. The Playground is for advanced work — multiple models, presets, aspect ratios, reference images, and a chat where you can describe changes in plain language — and is where you experiment before embedding the result.

Quality and how closely an image follows your prompt depend on the underlying model, so very complex or highly specific requests may not land on the first try. Refine your prompt and regenerate, or use a reference image in the Playground to steer the result closer to what you have in mind.