← Leaderboard

Large Behavior Model (LBM)

VLA
Toyota Research Institute · Jul 7, 2025 · built on Diffusion Policy

Multitask diffusion-policy model trained on about 1,700 hours of TRI and Open X-Embodiment data, studied with blind, randomized sim and real trials. TRI self-reports that fine-tuned LBMs beat single-task policies and reach their performance with 3–5x less task data, on TRI’s own tasks rather than a public benchmark.

Type
VLA
Parameters
—
Backbone
Diffusion Transformer on CLIP ViT image and text features
Weights
Closed
License
–
Commercial use
Unclear
Released
Jul 7, 2025
Robots
bimanual Franka FR3
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

Built on Large Behavior Model (LBM)