BEHAVIOR Challenge
An embodied intelligence benchmark based on the OmniGibson simulator that requires robots to complete 100 long-horizon household tasks in home environments, evaluating long-horizon planning and fine manipulation capabilities.
Timeline (last 90 days)
In the 2026 BEHAVIOR Challenge, an independent team used a single VLA model to tackle 100 long-horizon household tasks, ranking ninth on the leaderboard and being the only independent entrant in the top ten.
The BEHAVIOR Challenge requires robots to complete 100 household tasks in the OmniGibson simulator, each involving long-horizon planning and fine manipulation.
On the 50 tasks shared between 2025 and 2026, the model's Q value improved from 0.260 to 0.371; on all 100 tasks, the model scored 0.392.
The public submission included only part of the Astra results, with a final score of 0.446 Q and a full task success rate of 27.3%.
All Facts (4)
The BEHAVIOR Challenge requires robots to complete 100 household tasks in the OmniGibson simulator, each involving long-horizon planning and fine manipulation.
50%UnverifiedOn the 50 tasks shared between 2025 and 2026, the model's Q value improved from 0.260 to 0.371; on all 100 tasks, the model scored 0.392.
50%UnverifiedThe public submission included only part of the Astra results, with a final score of 0.446 Q and a full task success rate of 27.3%.
50%UnverifiedIn the 2026 BEHAVIOR Challenge, an independent team used a single VLA model to tackle 100 long-horizon household tasks, ranking ninth on the leaderboard and being the only independent entrant in the top ten.
50%