Microsoft Open Source Multimodal AI Agent - Magma

At 3am this morning, Microsoft opened sourced the multimodal AI Agent basic model - Magma on its official website. Compared with traditional agents, Magma has multimodal capabilities across digital and physical worlds, and can automatically process different types of data such as images, videos, and texts. For example, you can use Magma to automatically place e-commerce orders and check weather; It can also automatically operate physical robots or receive assistance while playing real chess. In addition, Magma also has built-in psychological prediction function, enhancing its understanding of spatiotemporal dynamics in future video frames, and accurately inferring the intentions and future behaviors of characters or objects in the video.