88Z AI, 18 August — HiDream Future has released HiDream-O1-World, a native fully multimodal interactive world model supporting text, image, and interactive inputs. It features roaming, editing, and interaction capabilities. Built on the proprietary UiT architecture, it achieves breakthroughs in spatiotemporal and physical consistency, topping the WBench Navi leaderboard on its debut. Applications include AI interactive movies and games, embodied AI simulation, and 3D scene production.
At a glance
- Built on proprietary UiT architecture, achieving breakthroughs in consistency and topping the WBench Navi leaderboard
- Supports multimodal inputs and roaming, editing, and interaction functions for AI media, embodied AI, and 3D production
The update
HiDream Future has officially launched HiDream-O1-World, a native fully multimodal interactive world model. The model supports various input modalities, including text, images, and interactions, with core functions centered on roaming, editing, and interaction. Technically, it utilizes the company's proprietary UiT architecture, achieving key breakthroughs in spatiotemporal and physical consistency. Due to its superior performance, the model topped the WBench interactive world model evaluation Navi sub-leaderboard in its first assessment. Its application scenarios currently span AI interactive movies and games, embodied AI simulation, and 3D scene production, providing robust technical support for these emerging industries.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- HiDream智象未来Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.