Search

Of all the things Google claims Gemini Omni can do, the most ambitious is also one that's hardest to verify from a demo reel alone. Google says that the model "combines an intuitive understanding of physics" with "Gemini's knowledge of history, science, and cultural context to bridge photorealism and meaningful storytelling." Testing that claim by asking the model to visualize a diagram I found in a textbook makes it far too easy. To stress-test this, I decided to rely exclusively on text-based prompts, giving the model nothing to work from except a written brief. For the first prompt, I asked