Robo Use

stanfordvl/behavior-1k

Household activities from the 2026 BEHAVIOR Challenge on a Galaxea R1Pro.

robouse run-many \
  --tasks tasks/behavior \
  --harness oracle \
  --out runs/behavior-oracle
About this dataset
Embodiment
Galaxea R1Pro mobile manipulator
Simulator
BEHAVIOR-1K (OmniGibson on Isaac Sim 5.1)
Requirements
NVIDIA RTX GPU · bench >=0.7.4,<0.8 · runs from a Robo Use checkout
Licence
Task packages Apache-2.0; adapted from BEHAVIOR-1K by Stanford Vision and Learning Lab, MIT
@misc{robouse_stanfordvl_behavior_1k_0_1,
  title        = {Robo Use stanfordvl/behavior-1k 0.1},
  author       = {{BenchFlow}},
  year         = {2026},
  howpublished = {\url{https://robouse.ai/hub/stanfordvl/behavior-1k/}},
  note         = {BenchFlow dataset, tag stanfordvl/behavior-1k-v0.1, commit f4872875a63e; adapted from BEHAVIOR-1K (Stanford Vision and Learning Lab)}
}

43 tasks

Filters

Task type

Preview
Details

↑ ↓ move between rows · Enter shows details · O opens the task

Each task is a package in the task format, pinned to robohub commit f4872875a63e. The verifier and the reference solution are not published.

Agent harnesses

Rank#Model · harnessSolve rateSolvedTasks run
1GLM-5.3 · Claude Code 100% 5 of 5 5 of 43
2GLM-5.3 · Codex 100% 5 of 5 5 of 43
3GLM-5.3 · mini-swe-agent 100% 5 of 5 5 of 43

Per task

38 tasks with no trials yet not shown.

✓ solved ✗ not solved ! infrastructure failure · not run