OpenAI Set to Launch GPT-5 in Early August After Multiple Delays

OpenAI is gearing up to release its highly anticipated GPT-5 model in early August 2025, marking the end of a series of delays that pushed the launch from May. The model is currently undergoing final testing, with CEO Sam Altman publicly confirming the imminent release and sharing his excitement about working with the system.

A Unified AI Powerhouse

GPT-5 is poised to integrate the reasoning capabilities of the o3 series, positioning it as OpenAI’s first fully unified model. Unlike previous iterations that required users to choose between specialized versions, GPT-5 promises a versatile, all-in-one system. Additionally, mini and nano versions will be available through the company’s API, catering to a range of user needs and computational demands.

Open Model Revival

In a significant move, OpenAI plans to release an open-source model by the end of July 2025—the first with publicly available weights since 2019. Sources describe this model as resembling the o3 mini, featuring robust reasoning capabilities. Although an earlier release was anticipated, Altman has attributed the delay to the company’s rigorous testing process, emphasizing the need to address potential issues before losing control over its usage post-launch.

---

Also read:

• OpenAI’s AI-Powered Browser: A Game-Changing Challenge to Google Chrome
• The Art of “Gradual Disclosure”: Is Jerome Powell’s Exit Being Carefully Orchestrated?
• Netflix and Disney Explore Runway AI’s Video Tools: A New Era for Hollywood or a Recipe for Controversy?

---

A Cautious Yet Bold Step Forward

The dual release strategy reflects OpenAI’s commitment to innovation while balancing safety and accessibility. GPT-5’s advanced features and the open-source model’s return signal a new chapter for the AI community, though the company’s cautious approach underscores its focus on mitigating risks. As the August launch nears, all eyes will be on how these developments shape the future of AI technology.