Gemini 4 Pro and Opus 5 both generate a three.js 3D dragon from the same prompt
With Opus 5 running at Max effort on the same prompt, Gemini 4 Pro still kept up, which the author sees as a sign it is now competitive for complex visual coding.
🚨 GEMINI 4 PRO IS GOING HEAD-TO-HEAD WITH OPUS 5. I gave both models the exact same prompt: “Create a 3D dragon using Three.js. Use a white background so the main object stays visible.” Opus 5 was running at Max effort. Gemini 4 Pro still managed to keep up. And that’s the interesting part. This isn’t a benchmark score. It’s the actual output side by side, built from the same prompt. Gemini 4 Pro is starting to look seriously competitive for complex visual coding. Which one would you pick?