{"id":"6ab6bd66-6794-4509-bb25-12479104d5a2","slug":"graspvla","name":"GraspVLA (Galbot)","description":"Galbot's grasping foundation model, GraspVLA (launched Jan 2025): an end-to-end Vision-Language-Action model pre-trained on SynGrasp-1B, a billion-frame synthetic grasping dataset with photorealistic rendering and domain randomization. Enables zero-shot generalization to new grasping tasks without additional training, powering dexterous manipulation on Galbot's robots.","brainType":"foundation-model","isOpen":false,"maturityStage":"research","architecture":null,"reviewStatus":"reviewed","sources":[{"url":"https://arxiv.org/abs/2505.03233","title":"GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data (arXiv)","sourceName":"arXiv (academic)"},{"url":"https://roboticsandautomationnews.com/2025/07/03/galbot-raises-151-million-to-scale-embodied-ai-humanoid-robots-partners-with-bosch-investment-arm/92864/","title":"Galbot raises $151M to scale embodied-AI humanoid robots","sourceName":"Robotics & Automation News"},{"url":"https://pku-epic.github.io/GraspVLA-web","title":"GraspVLA project page","sourceName":"PKU EPIC / Galbot"}],"keyFacts":[{"label":"Model type","value":"Vision-Language-Action model"},{"label":"Maturity stage","value":"research"},{"label":"Open source","value":"no"},{"label":"Training data","value":"SynGrasp-1B, a billion-frame synthetic grasping dataset with photorealistic rendering and domain randomization"},{"label":"Architecture","value":"end-to-end Vision-Language-Action model"},{"label":"SynGrasp-1B synthetic pretrain (primary arXiv)","value":"The GraspVLA arXiv paper (abs/2505.03233) from Galbot/PKU describes a grasping VLA pretrained entirely on SynGrasp-1B, a billion-frame synthetic grasping dataset with about 10,000 objects across 240 categories, photorealistic rendering, and domain randomization, plus Progressive Action Generation co-training with Internet semantics. Treat dataset scale as paper-stated (arXiv)"},{"label":"Zero-shot real-world grasp results (primary arXiv)","value":"The same paper reports real-world zero-shot grasping success around 90%+ across lighting, background, distractor, and height variants on both synthetic-category and web-category object sets, outperforming fine-tuned pi0 and OpenVLA baselines in the authors' tables. Treat rates as paper-stated experimental results (arXiv)"},{"label":"SynGrasp-1B generation pipeline","value":"The project page says SynGrasp-1B is built from more than 10,000 object meshes across 240 Objaverse categories, with stable grasps from BoDex and trajectories planned by CuRobo, then domain-randomized in simulation (GraspVLA)."},{"label":"CoRL 2025 release and labs","value":"The project page lists Galbot, Peking University, the University of Hong Kong, and the Beijing Academy of Artificial Intelligence, and marks the work as a CoRL 2025 paper that will release SynGrasp-1B and pretrained weights (GraspVLA)."}],"aliases":["GraspVLA"],"collisionRisk":"low","reviewNote":null,"builtOnBrainId":null,"createdAt":"2026-05-30T15:17:26.446Z","updatedAt":"2026-10-10T01:22:41.927Z","jsonLd":{"@context":"https://schema.org","@type":"SoftwareApplication","@id":"https://registry.deploy.report/brains/graspvla","url":"https://registry.deploy.report/brains/graspvla","name":"GraspVLA (Galbot)","alternateName":["GraspVLA"],"description":"Galbot's grasping foundation model, GraspVLA (launched Jan 2025): an end-to-end Vision-Language-Action model pre-trained on SynGrasp-1B, a billion-frame synthetic grasping dataset with photorealistic rendering and domain randomization. Enables zero-shot generalization to new grasping tasks without additional training, powering dexterous manipulation on Galbot's robots.","identifier":"6ab6bd66-6794-4509-bb25-12479104d5a2","applicationCategory":"foundation-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":3,"primary_source_types":["preprint"]}}}