Phi-3 models are the most capable and cost-effective SLMs, outperforming models of the same size and next size up across a variety of language, reasoning, coding, and math benchmarks.
Phi-3 models are trained using high-quality data, ensuring they are ready to use out-of-the-box.
Phi-3 models are instruction-tuned, meaning they are trained to follow different types of instructions reflecting how people normally communicate.
Phi-3 models were developed in accordance with the Microsoft Responsible AI Standard, ensuring they are responsibly developed, tested, and deployed.
Resource-constrained environments
Latency-bound scenarios
Cost-constrained use cases, particularly those with simpler tasks
Get started with Phi-3 on Microsoft Azure AI Studio
Explore the Phi-3 model on Hugging Face and Ollama
Use Phi-3 for resource-constrained environments, latency-bound scenarios, and cost-constrained use cases