Asked GPT-6 Astra and Claude Opus 5 to make a rice field in Three.js 🌾 Both set to Max thinking. Which one did it better?
115,629 次观看
👀 Since GLM-5.3-Flash supports vision... We gave the models a koi pond image as a reference and asked them to recreate it as an animation. Claude Opus 5, GLM-5.3-Flash, and Kimi K3 all followed the instruction really well, and the water movement they created is awesome too. Which one do you think did it best? 👇