← Leaderboard

Lumo-1

VLA
Astribot · Dec 9, 2025

Astribot’s reasoning VLA for its S1 robot, co-trained with data from other robots. Results are self-reported on Astribot’s own tasks (action accuracy in unseen environments rising from 86.98% to 92.95% in stage 2; stage-3 comparisons with π0 are in figures only).

Type
VLA
Parameters
—
Backbone
Qwen2.5-VL-7B
Weights
Unknown
License
–
Commercial use
Unclear
Released
Dec 9, 2025
Robots
Astribot S1
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