Microsoft releases lightweight AI model

On Tuesday, Microsoft unveiled a compact artificial intelligence model, aiming to appeal to a broader clientele with cost-effective options.

This latest version, named Phi-3-mini, marks the debut of the company's trio of small language models (SLMs), representing Microsoft's commitment to a technology poised to significantly impact global operations and work methodologies.

"Phi-3 is not slightly cheaper, it's dramatically cheaper, we're talking about a 10x cost difference compared to the other models out there with similar capabilities," said Sébastien Bubeck, Microsoft's vice president of GenAI research.

SLMs are tailored for executing simpler tasks, making them accessible to businesses operating with limited resources, Microsoft stated.

Phi-3-mini is now readily accessible through Microsoft's Azure cloud service platform, the AI model repository of Hugging Face, and Ollama, a framework facilitating model execution on local machines, according to the company's announcement.