Meta V-JEPA 2: Meta Introduces Advanced 1.2 Billion-Parameter World Model for Visual Understanding, Prediction for Robot Interaction With Unfamiliar Objects, Environment
Meta V-JEPA 2, a new advanced 1.2 billion-parameter world model designed for visual understanding and prediction. V-JEPA 2 is trained primarily on video, it enables robots to interact with unfamiliar objects and environments by imagining consequences before acting.
Meta AI launched a new world model, V-JEPA 2, that comes with state-of-the-art performance in the visual understanding and prediction of the physical world. The Meta Video Joint Embedding Predictive Architecture 2 (V-JEPA 2) is a 1.2 billion-parameter model that comes with improvements on the first model, V-JEPA, shared in 2022. The new model enables the robots to interact with unfamiliar (unseen) objects and environments in order to complete a task. Before taking action, the Meta V-JEPA 2 model, trained primarily on video, imagines the potential consequences like understanding, predicting and planning. OpenAI o3 Pro Model Now Available to ChatGPT Team, Comes With Advanced Reasoning, Visual Input Support and Web Search Capabilities.
Meta V-JEPA 2 Released, Trained on Video for Robots Interactions, Action Predictions
Introducing V-JEPA 2, a new world model with state-of-the-art performance in visual understanding and prediction.
V-JEPA 2 can enable zero-shot planning in robots—allowing them to plan and execute tasks in unfamiliar environments.
Download V-JEPA 2 and read our research paper… pic.twitter.com/Zdibfs8ZMQ
— AI at Meta (@AIatMeta) June 11, 2025
(The above story first appeared on LatestLY on Jun 12, 2025 01:19 PM IST. For more news and updates on politics, world, sports, entertainment and lifestyle, log on to our website latestly.com).
