OpenAI plans to bring a ChatGPT 5 version after it introduced GPT-4o just a few months back. Sam Altman, the founder of OpenAI posted on social media that GPT-4o is ready for another upgrade.
Sam Altman posted a photo of a Strawberry plant mentioning that he loves summer in the garden.
Decoding ‘project strawberry’
On Wednesday, Sam Altman shared a photo of some strawberries on X (previously known as Twitter), which is a social media platform, accompanied by a comment that said, “I love summer in the garden.” After he made the post, it went viral. Many users came up with their own explanation.
Some speculated that the X post might indicate an upgrade of GPT-4o that Sam Altman had been talking about last month. Several users made speculations that the upcoming project could be GPT-5 large language model (LLM). According to Mashable, this news was also confirmed by Bindu Reddy, CEO of open-source AI startup Abacus AI, supporting that Altman’s post was a “reference to Project Strawberry.”
According to Reuters, the ChatGPT-maker has been working on a model capable of advanced reasoning that’s one step closer to autonomous AI agents. Other leaks of internal documents explained Strawberry (formerly known as Q*) is a model that will allow “the company’s AI to not just generate answers to queries but to plan ahead enough to navigate the internet autonomously and reliably to perform what OpenAI terms ‘deep research.’”
What’s next
A month back Sam Altman had claimed that his upcoming ‘project Strawberry,’ aims to improve AI capabilities such as interpretation of data and analysing user’s pattern , focusing on more accurate info.
As reported by The Decoder, during the Aspen Ideas Festival, Altman spoke and expressed optimism about GPT-5. However, Altman had also said that there is still a lot of work to be done on the new model. He said that many errors have been found in the GPT-4o model and he aims to develop them.
OpenAI has also come up with a small yet powerful version of the GPT-4o. OpenAI claims that this GPT-4o mini version can give you superior performance in code and text compared to other versions. The company also claimed that this mini version is cheaper than its other models.
Furthermore, Altman said that, “A lot of the things that GPT 4o gets wrong, you know, can’t do much in the way of reasoning, sometimes just sort of totally goes off the rails and makes a dumb mistake, like even a six-year-old would never make.”
Follow FE Tech Bytes on Twitter, Instagram, LinkedIn, Facebook.