Odyssey has unveiled its latest advancement, Odyssey-2 Max, which is touted as its largest and most powerful world model to date. This model generates open-ended video simulations in real-time at 720p resolution, allowing it to interact dynamically based on user actions rather than simply producing pre-rendered sequences. Odyssey-2 Max represents a significant evolution in world model technology, building on previous releases like Odyssey-2 Pro, and pushes the boundaries of physical realism by incorporating fluid dynamics and enhancing real-time interactive capabilities.

Odyssey: Odyssey is an AI lab pioneering general-purpose world models, causal multimodal systems that predict and interact with the physical world over long horizons. The lab recently launched Odyssey-2 Max, its most powerful model yet, which streams real-time interactive 720p simulations that respond to user actions. This development advances the state-of-the-art in physical accuracy for world modeling.
odysseyml: @odysseyml is the official X account of Odyssey AI lab, a verified organization focused on multimodal intelligence through world models. The account announced the release of Odyssey-2 Max, emphasizing its role in moving beyond language models toward real-time world simulation. It engages the community with updates on interactive AI video capabilities.

Product: Odyssey-2 Max generates open-ended video simulations frame-by-frame without relying on game engines or pre-rendered assets.
Evolution: Follows recent releases like Odyssey-2 Pro, marking accelerated progress in general-purpose world model technology.
Advancement: The model pushes boundaries in physical realism, enabling fluid dynamics and interactive responses in real time.