According to multiple reports, OpenAI is preparing to launch its next-generation model, GPT-5, in early August 2025. While the company has not officially confirmed a date, CEO Sam Altman has publicly stated that the new model is “coming soon” and teased its advanced capabilities.
A significant change in GPT-5 is its architecture. It reportedly merges the standard GPT series with the “o-series,” which is specialized for logical reasoning. This aims to create a more streamlined user experience. Instead of requiring users to select a specific mode for complex tasks, GPT-5 will automatically determine when to apply advanced reasoning based on the query.
Performance and versions
Reports from sources like The Information suggest that GPT-5 shows major improvements in software development, excelling at complex academic and competitive programming challenges. Early testers have also noted its powerful reasoning abilities. In a podcast appearance, Sam Altman described an instance where the model perfectly answered a complex question that he could not.
GPT-5 is expected to be released in three versions:
- A standard version for ChatGPT and the API.
- A “mini” version for both platforms.
- A “nano” version available only through the API.
In parallel, Microsoft is reportedly testing a “Smart Mode” for its Copilot assistant. This feature would leverage GPT-5’s ability to switch between quick answers and in-depth analysis, preparing the tool for the new model’s launch.
Sources: The Decoder, The Verge, Livemint