Overview
- Image creation is free in the Gemini app, with reports of daily limits and some features consuming credits in tests.
- The figurines are generated by Gemini’s Flash image model, cited as 2.5 or 2.25, and are accessible through the Gemini app and Google AI Studio.
- Users worldwide, notably in India, are sharing results, with public figures such as Assam Chief Minister Himanta Biswa Sarma posting their miniatures.
- Hands, fine details and on‑image text remain recurring weak spots, with one test even misspelling “Happy” in a Comic Expressions collage.
- Google says outputs carry invisible SynthID watermarks and that user photos are not used for training without consent, while reports conflict on whether celebrity prompts are blocked.