{"id":"738c50ad-11cd-41d6-b5a4-1ac63e2acac2","slug":"rt-2","name":"RT-2 / RT-X","description":"Google DeepMind's foundational 2023 research VLA. RT-2 is a transformer VLA trained on web text and images that directly outputs robot actions. Instantiations on PaLM-E and PaLI-X. RT-2-X is a 55B-parameter variant. Chain-of-thought reasoning for long-horizon planning. Superseded by Gemini Robotics for commercial paths, but conceptually ancestral to GR00T, Helix, pi0, and the broader dual-system descendants.","brainType":"research-model","isOpen":false,"maturityStage":"research","architecture":"Transformer VLA trained on web text and images, directly outputs robot actions. Instantiations on PaLM-E and PaLI-X. RT-2-X is 55B parameters. Chain-of-thought reasoning for long-horizon planning. Trained on web data plus Open X-Embodiment.","reviewStatus":"reviewed","sources":[{"url":"https://deepmind.google/blog/rt-2-new-model-translates-vision-and-language-into-action/","label":"Google DeepMind Blog: RT-2 (primary)"},{"url":"https://blog.google/","label":"Google Blog: RT-2 coverage"},{"url":"https://arxiv.org/abs/2307.15818","title":"RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control (paper)","mediaType":"article","sourceName":"arXiv (Google DeepMind)"},{"url":"https://deepmind.google/blog/scaling-up-learning-across-many-different-robot-types/","title":"Scaling learning across robot types (Open X-Embodiment + RT-X)","mediaType":"article","sourceName":"Google DeepMind"},{"url":"https://arxiv.org/abs/2310.08864","title":"Open X-Embodiment: Robotic Learning Datasets and RT-X Models (paper)","mediaType":"article","sourceName":"arXiv (Open X-Embodiment)"},{"url":"https://research.google/blog/rt-1-robotics-transformer-for-real-world-control-at-scale/","title":"RT-1: Robotics Transformer for real-world control at scale (lineage)","mediaType":"article","sourceName":"Google Research"},{"url":"https://arxiv.org/abs/2212.06817","title":"RT-1: Robotics Transformer for Real-World Control at Scale (paper)","mediaType":"article","sourceName":"arXiv (Google)"}],"keyFacts":[{"label":"Significance","value":"First-of-its-kind VLA; demonstrated VLMs can become VLAs"},{"label":"Ancestry","value":"Conceptual ancestor of the GR00T / Helix / pi0 dual-system descendants"},{"label":"Powered platforms","value":"Research robot platforms; not a commercial-deployment brain"},{"label":"Successor","value":"Superseded by Gemini Robotics in DeepMind's commercial roadmap"}],"aliases":["RT-2","RT-X","RT-2-X"],"collisionRisk":"low","reviewNote":null,"builtOnBrainId":null,"createdAt":"2026-05-28T14:06:08.085Z","updatedAt":"2026-05-31T21:06:06.056Z","jsonLd":{"@context":"https://schema.org","@type":"SoftwareApplication","@id":"https://registry.deploy.report/brains/rt-2","url":"https://registry.deploy.report/brains/rt-2","name":"RT-2 / RT-X","alternateName":["RT-2","RT-X","RT-2-X"],"description":"Google DeepMind's foundational 2023 research VLA. RT-2 is a transformer VLA trained on web text and images that directly outputs robot actions. Instantiations on PaLM-E and PaLI-X. RT-2-X is a 55B-parameter variant. Chain-of-thought reasoning for long-horizon planning. Superseded by Gemini Robotics for commercial paths, but conceptually ancestral to GR00T, Helix, pi0, and the broader dual-system descendants.","identifier":"738c50ad-11cd-41d6-b5a4-1ac63e2acac2","applicationCategory":"research-model","publisher":{"@id":"https://deploy.report/#organization"}},"framework_metadata":{"framework_schema_version":"0.1.0","verification_status":"verified","maturity_stage":"research","lifecycle_state":null,"architectural_position":{"cohort":null,"sub_cohorts":[]},"within_cohort_verified_vs_claimed_pair":null,"cap_flags":[],"verification_depth":{"sources_count":7,"primary_source_types":["preprint"]}}}