OpenAI's 'Strawberry' AI model with enhanced reasoning set coming soon; here's all you need to know

OpenAI is reportedly gearing up for the release of its new AI model, codenamed "Strawberry," by the end of September. According to a report by The Information, this new model boasts enhanced reasoning capabilities and excels at solving complex mathematical problems.

Strawberry: Final Testing Phase and Potential Launch

Sources familiar with the development suggest that Strawberry is currently in the final stages of internal testing, indicating a public or beta release is imminent. The AI model could be launched within the next two weeks, although release timelines are subject to change.

It is also worth noting that "Strawberry" is an internal codename, and OpenAI might introduce the model with a different brand name at launch.

Enhanced Reasoning and "Thinking" Capabilities

Strawberry is reported to possess "the ability to think," demonstrating a more deliberate approach to problem-solving. Instead of instantly generating a response, the AI model takes 10-20 seconds to process the prompt and formulate an output, suggesting a more thoughtful and analytical process.

Availability within ChatGPT and via API

The new AI model will be integrated into OpenAI's popular chatbot, ChatGPT, making its advanced reasoning capabilities accessible to a wider audience. Additionally, Strawberry will be offered as a separate product, likely targeting clients who utilise OpenAI's API for specific applications.

At launch, Strawberry will focus solely on text-based tasks, without support for images or videos. However, OpenAI's ongoing development efforts suggest that future iterations could incorporate multimodal capabilities, expanding its potential applications.

Not GPT-5, but a Separate Model

It's important to note that Strawberry is not the next generation of OpenAI's GPT language model (GPT-5). It is a separate model designed to address specific challenges related to reasoning and problem-solving.