XPENG Unveils The “World Model Accelerator” X-Cache, Which Requires No Training, Is Plug-And-Play, And Boosts Inference Speed By 2.7 Times
AI-summarised brief · reviewed before publication
XPENG unveiled the X-Cache, a world model accelerator that boosts inference speed by 2.7 times without requiring training. It achieves this by identifying reusable image regions and reducing redundant computations. The X-Cache can be directly applied to world models in a fast and lightweight manner, enhancing efficiency and reducing resource consumption. This technology has been successfully deployed in XPENG's autonomous driving world model, X-World.