← Leaderboard

AutoRT

Agent system
Google DeepMind · Jan 4, 2024

Orchestration system in which a VLM describes the scene and an LLM proposes and filters tasks for a fleet of robots running RT-1 or RT-2, under safety prompts and human supervision. Over seven months it ran up to 20 robots at once (52 in total) and collected 77,000 real episodes across 6,650 unique tasks; it is a data-collection system with no benchmark scores.

Type
Agent system
Parameters
—
Backbone
–
Weights
Closed
License
–
Commercial use
Unclear
Released
Jan 4, 2024
Robots
Everyday Robots mobile manipulator (Google Robot)
SO-101
–
LeRobot
–
Inputs
–
VRAM
–

Benchmark results

0 of 13 benchmarks

No benchmark results recorded yet.

Not reported on 13 benchmarks
Wrong or missing number? Report an error