Embodied Analysis
Unified evaluation and benchmarks for embodied AI world models and physical agents.
Is this your creation?
Description
A benchmarking platform that provides unified evaluations for embodied AI, comparing physical agents and world models across simulation and real-robot tasks. It uses human pairwise judgments to aggregate performance into Elo ratings, assessing dimensions from physical plausibility to goal completion. The evaluations include models such as Gemini 4 Argon, Claude Opus 5.5, and Muse Spark 1.3.
Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.