- OpenAI is reportedly gearing up to integrate Strawberry into ChatGPT to improve its reasoning capability. It could launch as early as this fall.
- The Strawberry engine can also solve complex math and programming problems, perform in-depth research, plan, and generate high-quality synthetic data.
- OpenAI is also training the successor of GPT-4/4o called 'Orion' which could be the much-awaited GPT-5 model.
Earlier, we reported that OpenAI is working on a new AI engine called Project Strawberry. It aims to bring human-level reasoning capability to ChatGPT and can solve complex math and programming problems. Now, a report by The Information confirms that OpenAI is indeed gearing up to integrate the Strawberry engine into ChatGPT.
According to two sources familiar with the development, OpenAI may launch Strawberry to ChatGPT this fall. The Information has learned that the hot AI startup has already demoed Strawberry to US national security officials. The report further says that Strawberry was able to solve the New York Times “Connections” word puzzle in an internal demo.
Apart from that, Strawberry can also perform in-depth research, plan, and perform actions, unlocking agentic capability. In our earlier piece, we noted that Strawberry’s approach is very similar to a technique called STaR (Self-Taught Reasoner), proposed by Stanford researchers that can significantly improve reasoning capability.
Apart from that, OpenAI is reportedly working on its next-generation AI model codenamed ‘Orion.’ It’s said to be the successor of GPT-4/4o and could be the next GPT-5 model. Moreover, Strawberry is helping in generating high-quality synthetic data for Orion to be trained on. The release date for Orion is not known yet.
