Embodied Playground tests LLM agents on 410 simulated physical reasoning tasks

Can an LLM agent play Angry Birds?

Introducing Embodied Playground 🤖
410 simulated task environments that test physical, strategic, and spatial reasoning.

🌏 Website: https://winniechen2002.github.io/Agent-bench-website/
📄 Paper: https://arxiv.org/abs/2610.27039
💻 Code: https://github.com/Winniechen2002/Embodied-Playground