OpenAI may soon release a new artificial intelligence model under the working name Strawberry. Rumors about this model have been circulating for some time, but now there is a possibility that it will be released within two weeks. This was reported by Android Authority.

Strawberry will focus on reasoned responses and “deep research”. Unlike other OpenAI models that try to provide an answer as quickly as possible, this model will take 10 to 20 seconds to respond. This approach will allow AI to reduce the number of errors and provide more reasoned and complete answers to tasks that require complex or multi-step reasoning.

The new model is expected to be used in areas such as mathematical problem solving, coding, and business-related tasks such as developing detailed marketing strategies.

It is reported that two people were able to test Strawberry and shared their experience of using it. First of all, the new model supported only text communication during testing. It also took a long time to answer even the simplest questions, and the function of remembering previous conversations was unstable.

The new model from OpenAI will be integrated into ChatGPT, but as a separate option. This model will also have different prices. It is expected that users can be limited in the number of requests per hour, and with a higher subscription level, this number will be increased, as well as the speed of responses. However, at the beginning, the new model is likely to be available as part of the ChatGPT Plus plan.

We would like to remind you that OpenAI’s new AI model was actively discussed in July. Back then, it was reported that the principles of the model were kept secret even within the company. Recently, it was also reported that OpenAI already has more than 1 million business users and plans to launch new subscription tiers, possibly including access to Strawberry.