"OpenAI's GPT-5.2 Pro does better at solving sophisticated math problems than older versions of the company's top large language model, according to a new study by Epoch AI, a non-profit research institute."
"GPT-5.2 Pro solved four problems that had been too difficult for any other AI models to solve, and of the 13 problems that any other model had previously solved, it was able to solve 11, Epoch reported."
"This means GPT-5.2 Pro had solved 31% of Epoch AI's challenges, a rise from the previous score best of 19%."
Epoch AI, a non-profit research institute, evaluated advanced math problems to benchmark large language models. GPT-5.2 Pro solved four problems that no other AI had solved previously and solved 11 of the 13 problems that other models had managed, reaching success on 31% of the challenges. The 31% result represents an improvement over the previous top score of 19%. These outcomes indicate a measurable advancement in GPT-5.2 Pro's ability to handle sophisticated mathematical problem-solving compared with older versions of the model.
Read at Computerworld
Unable to calculate read time
Collection
[
|
...
]