OpenAI Unveils ‘Strawberry’: New Reasoning-Focused AI for ChatGPT

OpenAI is going to release “Strawberry,” a new AI model for ChatGPT that focuses on reasoning. It will give more thoughtful and accurate answers to difficult questions, but it will take longer to process.

OpenAI has said that in two weeks, there will be a new artificial intelligence-based model named “Strawberry.” This fresh AI model is to sharpen the reasoning faculties of ChatGPT to enhance its responses to tricky questions. Obviously, Strawberry is not the typical AI model since it ‘thinks’ about problems before’solving’ them. It focuses on providing solutions to problems with a concise and thoughtful approach.

What Makes Strawberry Different?

Strawberry is not just the same as ChatGPT but with a different name. This is the type of AI that is designed to solve problems using arguments and logic. It is an AI that is meant for tackling challenging tasks that require a higher level of problem-solving skills. It’s unlike other models of AI since this particular model takes time before responding. Other earlier AI models always preferred to prioritise velocity and not accuracy. This delay allows Strawberry to provide more comprehensive and factual responses in comparing to risks since, in some cases, solving a problem may involve more than one step, like in mathematical problems, business strategies and planning, and so on.

There is a fact that many AI models are improving on multimodal tasks such as input and output handling both text and images. Comparatively, Strawberry will be capable only of text-based conversational ability with the characters. In this way, it wants to reduce the risk of such inputs providing incorrect or what is known as ‘hallucinated’ data, which AI models have to think about trying to understand what an image or other data signifies. It is worth mentioning that Strawberry is more suitable for those users who are searching for detailed and accurate answers since it provides only text.

A rather large difference associated with strawberries is the duration taken in the processing stage. It is also said that it will take exactly twenty seconds for the Strawberry to answer, at least as the creators of the current variant of ChatGPT in the process. This waiting might appear as a disadvantage; however, it is because the model is capable of thinking more deeply than any person. The AI uses relatively more time to analyze the data so that it gives a relevant and more appropriate solution to individuals who are posing complex questions.

Initial trials of Strawberry suggested that the drawback to this is the delay in receiving a response compared with other platforms, although the responses are slightly better quality. This is beneficial for people who require basic study because that may take longer and will be a bit better off compared to the outcome.

As pointed out earlier on the fact that Strawberry has more enhanced features than its predecessors, then it might demand more computational capacity than the others. It could mean that it is expensive in some way, and it may be that in sending text, for example, one can only send a few per hour. There are also the working theories that OpenAI is planning to expand its pricing options and create a paid tier. End users willing to pay more might be able to have a quicker response rate and more use potential.

What OpenAI has done with Strawberry is just one of the many attempts they’re making to enhance the technology in this world. The company has also been engaging in the development of Orion and other artificial intelligence models and new versions. Here, the CEO of OpenAI Japan, Tadao Nagasaki, stated at the KDDI Summit 2024 that GPT-Next, which is yet to be released, will be a hundred times more robust than GPT-4.

There is a plan that GPT-Next will enhance from what has been created in Project Strawberry, faster, more efficiently, and highlighted for bigger and more complex tasks. I think that working on artificial general intelligence, OpenAI already got further, and Strawberry is the next step to look for what AI is capable of.

What to Expect for Users

This is true because Strawberry is an interesting development in AI technology for the user. It may not be the best tool to use if you want to answer trivial questions because it is designed to be very good at complex, iterative, and time-consuming calculations. This is because Strawberry is going to be a better, more thorough, and well thought out answer, whether you are trying to answer complex math problems, organize a marketing plan, or get the right answer.

It seems that a new product is being developed in OpenAI all the time, and the release of Strawberry is a milestone in the history of AI. This new model, largely concentrating on reasoning and problem-solving approaches, will very soon become one of the most valuable for users of the ChatGPT tool in order to optimize its usage. More news is coming out, so to speak, when OpenAI is soon to release Strawberry in a few weeks from now.