Microsoft’s long-standing partnership with ChatGPT OpenAI has resulted in the company getting a leg up on its competitors like Google, Amazon and others in the manufacturing AI industry. However, the company is reportedly going it alone in creating a major new language model that could eventually compete with ChatGPT.

as per the informationVia unnamed sources, Microsoft’s in-house LLM is called MAI-1 (presumably the “M” stands for “Microsoft). It is developed with 500 billion parameters, compared to OpenAI’s GPT-4 which has over 1 trillion parameters.

The report says MAI-1’s development is being led by Microsoft’s new AI chief, Mustafa Suleiman. He was hired by the company in March after founding Google’s DeepMind AI division, and later founding his own AI startup Inflection.

Before being hired by Microsoft, Soliman worked on another LLM for Inflection. The Information reports that this new MAI-1 LLM he’s supervising doesn’t share any code with his previous work with Inflection, but he may use some training data from the same startup.

Microsoft probably wants to be less dependent on OpenAI for help creating AI services like its various Copilot versions, all of which currently use some version of ChatGPT.

Today’s report does not indicate which services will be used in the upcoming MAI-1 LLM. The information suggests that Microsoft may reveal more about this project at its annual Build developer conference, which kicks off in Seattle on May 21.

Microsoft has already released a number of smaller and lighter open source LLMs. Last week, it released the new Phi-3 family of small LLMs. They work with as low as 3.8 billion parameters. They are also designed to run on native devices such as smartphones, but may still handle advanced natural language processing for a wide variety of tasks.