VLA models

Octo

United States

First fully open-source generalist robot policy (data, weights and training pipeline); its compact diffusion-head transformer became a standard academic baseline for cross-embodiment manipulation.

Updated 2026-07-09 · Last verified 2026-07-09

OrganizationUC Berkeley / Stanford / CMU / Google DeepMind (academic consortium)
CountryUS
Release2024-05
Params (B)0.1
Open weightsyes
Architecturetransformer policy with diffusion action head; flexible observation and action spaces (Octo-Small 27M, Octo-Base 93M)
LicenseMIT
Modalitiesvision, language, action
Embodimentsmanipulator, cross-embodiment
Training data800k robot trajectories from the Open X-Embodiment dataset

Sources : arXiv (2024-05-20)Octo project (2026-07-09)GitHub (2026-07-09)

← Back to comparator · VLA models
Christian Verbrugge

Christian Verbrugge · D·Fairy

Bringing a physical AI solution into European industry?

15 years at KUKA on automotive OEM programs (€100M), now focused on edge AI and autonomous agents. I support robotics and embodied AI companies on their EMEA market entry: OEM and Tier 1 access, co-funded pilots, AI Act readiness.

Book an intro call