OpenAI’s GPT 5 set to launch in early August with advanced reasoning and mini versions: Report

OpenAI is preparing to release its next major AI model, GPT-5, as early as August, according to a report by The Verge. The model is expected to arrive alongside “mini” and “nano” versions available through API access, significantly expanding OpenAI’s suite of tools for developers and enterprises.

Microsoft engineers have reportedly been working on readying server capacity since May, with the initial timeline delayed due to further testing. OpenAI has yet to confirm an official release date, but CEO Sam Altman hinted at the model’s arrival in a podcast this week, saying, “We are releasing GPT-5 soon.”

Altman recounted a personal moment while testing the model. “I put it in the model, this is GPT-5, and it answered it perfectly,” he said during an appearance on This Past Weekend with Theo Von. “It was a weird feeling,” Altman added. “I felt useless relative to the AI.”

The Verge reported that the main GPT-5 model will feature integrated reasoning capabilities previously seen in OpenAI’s o-series models, particularly o3. This integration is intended to simplify the user experience in ChatGPT by eliminating the need to choose between model types. The combined system will also be made available via API. While the mini version will be accessible both on ChatGPT and through API, the nano version will be limited to API access.

The release of GPT-5 marks a step closer to what OpenAI has described as artificial general intelligence (AGI). Altman has previously said the model would not reach “gold level of capability” immediately after launch, but the unification of model families is part of OpenAI’s broader strategy to streamline AI use cases and performance.

The declaration of AGI remains a high-stakes moment for OpenAI. Under the current agreement, Microsoft would forfeit its rights to future revenue and model access from OpenAI upon a formal AGI milestone. Both companies have been in negotiations recently over adjustments to their partnership, with OpenAI seeking to restructure part of its business into a for-profit entity.

Meanwhile, another model release is imminent. OpenAI is reportedly preparing to launch an open-weight language model before the end of July. Sources say this model resembles o3 mini and will include advanced reasoning capabilities. It will be the first open-weight release by OpenAI since GPT-2 in 2019, and will be available on platforms like Azure, Hugging Face, and other cloud providers.

Altman confirmed a brief delay in the release of this open model, citing safety tests and a review of high-risk areas. But insiders say OpenAI is pushing to ship the model before GPT-5, possibly as a strategic move ahead of competitor announcements.