The unified episode format helps data loading across contributors, but observation spaces, actions and collection practices still vary. A training mixture must record which constituent datasets it used.
google-deepmind · Publication date not disclosed · Source accessed 2026-09-28
Official README inspected live. Repository code terms do not automatically cover model weights, data, or third-party assets.
Integration decision
The repository links dataset-specific information and citations. These are necessary for tracing provenance and assessing whether a particular subset fits a planned training use.
google-deepmind · Publication date not disclosed · Source accessed 2026-09-28
Official README inspected live. Repository code terms do not automatically cover model weights, data, or third-party assets.
Operational and reuse limits
Repository software and documents have stated licenses; each constituent dataset keeps its own terms. Do not apply one blanket license to the collection.
google-deepmind · Publication date not disclosed · Source accessed 2026-09-28
Official README inspected live. Repository code terms do not automatically cover model weights, data, or third-party assets.
What this evidence does not establish
Repository software and documents have stated licenses; each constituent dataset keeps its own terms. Do not apply one blanket license to the collection.
Related reading is an editorial crosslink. Sourced connections describe relationships reported in the cited material. A link to a versioned profile does not establish compatibility with that version unless the connection note explicitly identifies it.
Record history & verified changes
A research review records when we checked a source. It does not mark a product launch or a new deployment.
Initial reviewed record. No subsequent field change has been recorded.
google-deepmind · Repository · Publication date not disclosed · Source accessed 2026-09-28
License: Apache-2.0 repository software; CC BY 4.0 other repository content; constituent dataset licenses separate. Factual summary and attribution only; no upstream prose, images, weights, or dataset redistributed.
Official README inspected live. Repository code terms do not automatically cover model weights, data, or third-party assets.