RoboColosseum
AI overview
Sign in and the AI will write an overview from our coverage.
Headlines · 3
- DJiafei Claims GPT-6.2 Astra Tops RoboColosseum SO-101 Elo Rankings
DJiafei said GPT-6.2 Astra reached No. 1 in RoboColosseum's SO-101 Elo rankings after hundreds of same-day head-to-head comparisons in the wild, according to a post that included a robotics demonstration video.
Greg Brockman · OpenAI President · 🔥 4 - RoboColosseum launches open real-world robot evaluation
Hugging Face announced that it is a founding partner of RoboColosseum, an open evaluation initiative for robot models. RoboColosseum says participants can access more than 100 physical robots, including Unitree G1 and LeRobot SO-101 systems, across more than 140 real-world scenes, with rollout data—including failures—published on the Hugging Face Hub.
Hugging Face (X) · 🔥 0 - RoboColosseum Evaluates Robot Policies Across Distributed Sites
The paper introduces RoboColosseum, a distributed framework that combines unified policy serving with standardized evaluation protocols across locally operated robots in diverse environments. Its initial study evaluated five vision-language-action policies and three frontier AI agents across four embodiments, including SO-101, Franka, Bimanual-YAM, and G1 Humanoid, through 5,699 pairwise comparisons at 124 sites. The study reports weak ranking agreement across embodiments and substantial rank reversals between pre- and post-fine-tuning evaluations.
Hugging Face · Papers · 🔥 0
Experience and discussion from the community
Share my RoboColosseum experienceAsk about RoboColosseum
Nobody has shared their experience with RoboColosseum yet.