Octo Model Team: adaptable cross-embodiment diffusion policy resource

What changed

Combined flexible observation/task tokenization, block-structured transformer processing and diffusion action decoding into an open policy initialization adaptable to new sensors, action spaces and robots.

Scientific scope

Combined flexible observation/task tokenization, block-structured transformer processing and diffusion action decoding into an open policy initialization adaptable to new sensors, action spaces and robots. The method builds on transformers, action chunks, diffusion decoding and Open X-Embodiment data. Adaptation still requires target data and appropriate adapters; the reported results do not establish universal robot compatibility.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The method builds on transformers, action chunks, diffusion decoding and Open X-Embodiment data. Adaptation still requires target data and appropriate adapters; the reported results do not establish universal robot compatibility.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Octo: An Open-Source Generalist Robot Policy
• Octo Model Team: adaptable cross-embodiment diffusion policy resource — original project

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction