TLDR Tech / Launches
Artificial Analysis Releases AA-Robotics v0.1 for Embodied AI Testing
Artificial Analysis has unveiled AA-Robotics v0.1, a research preview for assessing advanced language models' capabilities in understanding, planning, and interacting with the physical world via a robot arm. Tests on eight common tabletop tasks revealed that leading language models, when integrated with coding agents, can control robot arms, but current speeds and costs make them impractical for widespread use. GPT-6 Astra achieved the highest score at 69, followed by GPT-6.1 Sol, Gemini 4 Argon, Claude Opus 5.5, and Claude Fable 5.1.
Read the full story