Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Microsoft just introduced Magma, a new artificial intelligence model Designed to help robots see what they see, understand and smarter. Unlike traditional AI Models, Magma processes various data at a time – with an effort to make a great leap to Microsoft “Agentic AI” or systems that can plan and execute tasks on behalf of the user.
The model, which uses the combination of vision and language processing, is taught videos, pictures, robotics, information and interface interactions to make more versatility than previous models.
The Microsoft Research Group on Github can manipulate Magman’s robots and how it can manage the user interfaces, how the user can fulfill the assignments.
The company partnered with the researchers of the University of Maryland University, Wisconsin-Madison and Washington University to develop technology.
Start comes as Tech Giants race to develop AI agents that can automate more aspects of daily life. Google develops robotics-oriented language models Openai’s operator tool To book, order groceries and order the forms to order and scroll through the specialized browser, designed to manage secular tasks such as ordering and filling out forms.
Jianwei Yang, a leading Microsoft researcher for the project, said CNET’s future for the future of the EU and the development of multimodal foundation for Chatbots.
“We believe that the next important step for the AI is the next important step for the development of developing agents that can not immediately understand and interact with both digital and physical environment.”
He said that Magman’s importance is the ability to deprive the gap for Multimodal AI agents because traditional AI models are superior in oral intelligence, but often fought in planning and real world action.
“Today, Robots often rely on specific exercises for special information on specific information, resulting in a limited ability to make generalized simple work in new positions and environments,” he said. “MAGMA changes this by significantly increasing the verbal and space intelligence, which allows you to perform robots at the top of the environment, or physically or physically or physically or physically or effectively.”
Meanwhile, Craig Le Clair – Author of the main analytical and random automation in Forrester – Sunday research information with the forecast, information on the combination of 25% of the 2025 robotics projects. He said that this announcement and others are the real turning point or a greater gift continues.
“Microsoft has given an important developer, but now you need to demonstrate the leadership in the productive and safe human-robot interaction,” said Le Clair.
