OpenAI is set to unveil its next-generation artificial intelligence model, GPT-5, in early August 2025, according to a report by The Verge. The release will include not only the flagship GPT-5 model but also two lightweight versions: GPT-5 Mini and GPT-5 Nano.
With this launch, OpenAI aims to unify its existing GPT and “o-series” models, such as the o3 engine, into a streamlined system. This integrated approach is expected to simplify user experience and elevate performance across reasoning, creativity, and task execution.
According to internal sources, the GPT-5 base and mini models will be available through both ChatGPT and OpenAI’s API, while the Nano version will be exclusive to API users. The Nano variant is designed for developers and businesses seeking fast, lightweight AI solutions.
In a recent statement, OpenAI CEO Sam Altman described the new model’s capabilities as “unsettlingly powerful,” noting how GPT-5 performed complex tasks effortlessly during internal testing-so much so that it made him feel “useless” compared to the AI.
OpenAI will also release an open-weight AI model later this month, said to be similar in performance to the current o3-mini. This version will be hosted on platforms like Azure and Hugging Face, enabling broader access for researchers and developers.
Free users of ChatGPT will continue to receive access to standard capabilities, while ChatGPT Plus and Pro subscribers are expected to gain access to more powerful features offered by GPT-5.
The new release is seen as a strategic step to maintain OpenAI’s lead in the rapidly evolving AI landscape, competing with the likes of Anthropic’s Claude, Google’s Gemini, and Meta’s Llama models.
