Microsoft Launches Small AI Models Family Phi-3

Recognizing that bigger isn't always better, Microsoft has released its smallest open AI models to date, Phi-3.

The new micro models "are the most capable and cost-effective small language models (SLMs) available," according to Microsoft. With its small size, the company said the models can fit on a smartphone and can be used, even if not connected to the Internet.

"What we're going to start to see is not a shift from large to small, but a shift from a singular category of models to a portfolio of models where customers get the ability to make a decision on what is the best model for their scenario," said Sonali Yadav, principal product manager for Generative AI at Microsoft.

The first official model in the Phi-3 family is the Phi-3-mini, a 3.8 billion language model that will be available in length variants of 4K and 128K tokens. Microsoft said it will be the first of its kind to support an extensive context window of up to 128K tokens without compromising on quality. Its dual availability allows users to choose the variant that best suits their computational needs and application requirements, ensuring versatility across a broad range of AI-driven tasks.

In the company's own head-to-head comparison tests, it found that Phi-3-mini outperforms models twice its size.

Phi-3-mini is designed with instruction tuning, which enables the model to interpret and execute a variety of instructions that mirror natural human communication. Microsoft said this will enhance the user experience by making the model highly intuitive and ready to implement right out of the box. By aligning closely with human instructive patterns, Phi-3-mini facilitates a smoother interaction between humans and machines for more natural conversations.

To accommodate a wide range of deployment environments, Phi-3-mini is integrated with a handful of Microsoft services and products. It is available through Azure AI and can also be run locally on developers' laptops via Ollama. Additionally, the model has been optimized for ONNX Runtime and supports Windows DirectML, providing cross-platform compatibility across GPUs, CPUs, and mobile devices. Moreover, Phi-3-mini can be deployed as an NVIDIA NIM microservice, featuring a standardized API interface that is optimized for NVIDIA GPUs.

Microsoft said that in the coming weeks, Phi-3-mini will be joined by Phi-3-small (7 billion language model) and Phi-3-medium (14 billion language model). The two upcoming larger micro models are said to be able to outperform OpenAI's GPT 3.5 Turbo at much smaller sizes.

As with all of Microsoft's language models, the company said it has prioritized safety and privacy in the development of the Phi-3 family. "Phi-3 models underwent rigorous safety measurement and evaluation, red-teaming, sensitive use review and adherence to security guidance to help ensure that these models are responsibly developed, tested, and deployed in alignment with Microsoft’s standards and best practices," wrote Microsoft.

About the Author

Chris Paoli (@ChrisPaoli5) is the associate editor for Converge360.


comments powered by Disqus

Subscribe on YouTube