OpenAI GPT 5 Is Coming Sooner Than You Think

Elyse Betters Picaro / ZDNET

Just when you thought OpenAI might take a breather after launching its ambitious ChatGPT Agent, the AI giant is stirring up excitement yet again. Whispers and reports are pointing towards an even bigger release on the horizon: the highly anticipated GPT-5.

An August Surprise The GPT 5 Release Date Nears

While talk about GPT-5 has been circulating for a while, recent developments suggest its launch may be just around the corner. OpenAI CEO Sam Altman himself teased that GPT-5 was "releasing soon" on X earlier this week. Adding fuel to the fire, a new report from The Verge, based on insider information, has pegged the potential release for early August. As is typical with OpenAI launches, this major model is expected to be accompanied by mini and nano versions, which will also be available for developers through an API.

The Road to GPT 5 A Model Evolution Timeline

OpenAI is due for a next-generation model. Let's take a quick trip back in time. When ChatGPT exploded in popularity in November 2022, it ran on GPT-3.5. In March 2023, GPT-4 arrived, marking a significant leap in intelligence. Then, March 2024 brought us GPT-4o, which improved on its predecessor and introduced powerful multimodal capabilities. Now, over a year later, the stage is set for GPT-5.

This wasn't always the exact schedule. In February, Sam Altman shared a roadmap on X, suggesting a release was just "weeks/months" away. However, plans changed. In April, Altman announced a pivot to release the o3 and o4-mini reasoning models first, pushing the GPT-5 launch back by a couple of months. The delay was attributed to the complexities of integrating all the different elements into the new flagship model.

What New Capabilities Will GPT 5 Bring

So, what's the big deal with GPT-5? According to Altman's initial roadmap, it represents a major architectural shift. GPT-5 is designed to combine the deep reasoning capabilities of the o-series models (like o3 and o4) with the powerful generative strengths of the GPT-series. The system will be smart enough to know when to apply which model's strength, giving users an optimal balance of speed, cost, and quality.

This is a huge win for users. While reasoning models excel at producing high-quality answers for complex problems, they are slower and more expensive. Their full power isn't always needed for simpler tasks. Conversely, defaulting to standard GPT models means missing out on superior responses for tasks that demand deeper thought. GPT-5 aims to automate this choice, ensuring the best tool is always used for the job.

Access Tiers and Managing Expectations

Altman's roadmap also indicated that GPT-5's power will be distributed across user tiers. Free users will reportedly get unlimited chat access to GPT-5 at a standard intelligence setting. Plus subscribers will be able to access a higher level of intelligence, while Pro subscribers will get the highest level available. The new models are expected to integrate all of ChatGPT's best features, including voice, search, and deep research tools.

However, it's wise to table expectations slightly. Altman recently reminded everyone that the model will be experimental and incorporate various research techniques. He clarified that it won't be at the IMO gold level of mathematical reasoning, a feat achieved by a different, still-anonymous model set for a later release. In another exciting development, OpenAI has also said that an open-source model is in the works, which would be its first since GPT-2.