GPT-4.1 Models Now Available on Microsoft Azure OpenAI Service and GitHub

Microsoft is announcing the availability of GPT-4.1, 4.1-mini, and 4.1-nano on Microsoft Azure OpenAI Service and GitHub, bringing in the latest iteration of the GPT model series. Capable of transforming applications, these models deliver a variety of enhancements that improve performance, efficiency, and versatility.

The GPT-4.1 models are engineered to excel at complex technical, coding, and instruction-following tasks, further poised to improve agentic workflows and enhance developer productivity. This iteration of the GPT-4o model delivers cleaner, simpler front-end code, according to Microsoft, paired with the capacity to accurately identify points of improvement in existing code. The code generated by these models consistently compiles and runs successfully, Microsoft noted.

Additionally, the GPT-4.1 models support longer contexts—up to one million token inputs. The longer the context window, the greater the model's ability to process and understand extensive context in a single interaction. This is well suited for complex, multi-step interactions, which often require more detail and nuanced understanding.

Finally, improved instruction enables the family of GPT-4.1 models to better carry out multi-step, detailed tasks. These models are also more intuitive and collaborative, increasing its flexibility for a broad range of applications.

Notably, GPT-4.1 maintains the same API capabilities as the GPT-4o model family—including tool calling and structured outputs.

Microsoft anticipates its release of supervised fine-tuning for GPT-4.1 and 4.1-mini, which will be available soon. This feature enables developers to better tailor the models according to their business needs, customizing the base models according to an enterprise's unique datasets, tone, domain terminology, and task workflows.

To learn more about Microsoft’s latest announcement, please visit https://www.microsoft.com/en-us/.