DeepSeek has released DeepSeek-OCR 2, AI can "see" an image in the same logical order as humans.

DeepSeek has released the new DeepSeek-OCR 2 model, which utilizes the innovative DeepEncoder V2 method, enabling AI to dynamically rearrange the various parts of an image based on its meaning, rather than mechanically scanning from left to right. This approach simulates the logical process followed by humans when viewing a scene. Ultimately, this model outperforms traditional visual-language models when processing complex layout images, achieving a more intelligent and causally reasoning visual understanding.