← Library · Frontier
Embodied-Reasoner Outperforms OpenAI O1 in Embodied AI Tasks
A joint team from Zhejiang University, Chinese Academy of Sciences, and Alibaba Damo Academy has open-sourced Embodied-Reasoner, a multimodal embodied reasoning model. It achieved an 80.96% task success rate in the AI2-THOR simulator, surpassing OpenAI o1 (71.73%), o3-mini (56.55%), and Claude-3.7 (67.70%). The model uses a three-stage training pipeline involving imitation learning, self-exploration, and self-correction.
Why it matters
Embodied-Reasoner demonstrates superior performance in complex interactive physical tasks, highlighting advancements in AI's ability to reason, search, and correct errors in simulated and real-world environments.
Learn one new AI thing every day.
Daily Deck sends you seven plain-English cards like this every morning. Free.
Start free