Microsoft is stepping up preparing server capacity to welcome the release of GPT-4.5 and GPT-5 models

According to reports, Microsoft engineers are going all out to prepare server capacity for OpenAI's upcoming GPT-4.5 and GPT-5 models. OpenAI CEO Sam Altman recently confirmed that GPT-4.5 will be released in a few weeks, and Microsoft is expected to launch this new AI model next week.

GPT-4.5, known as Orion, will be the next cutting-edge model for OpenAI and the company's last non-thinking chain model. Although OpenAI has revealed that the GPT-4.5 will have much more powerful capabilities than the previous generation of GPT-4, their eyes are also on the GPT-5 model, which will contain more major technological innovations. Under current plans, Microsoft expects the GPT-5 to be launched by the end of May this year, in line with the claim that Altman had previously promised that the next-generation model would be launched within a few months. However, release time may change due to plan adjustments.

Initially, OpenAI planned to launch GPT-4.5 by the end of 2024, but this plan has been postponed until early 2025. The GPT-5 is considered the more important of these two models, and Artman calls it "a system that integrates many of our technologies."

In addition, GPT-5 will integrate OpenAI's new o3 inference model, which debuted in a Christmas announcement last December. While OpenAI released the o3-mini last month, they no longer intend to launch the o3 as a standalone model, but instead integrate it with the GPT-5 system. This integration is in line with OpenAI's goal, to combine large language models to ultimately create a more powerful model that can be called universal artificial intelligence (AGI).