OpenAI says GPT-5.2 sets new state of the art on FrontierMath
GPT-5.2 Thinking solved 40.3% of FrontierMath problems, a new state of the art. GPT-5.2 Pro hit 93.2% on GPQA Diamond. OpenAI still calls such systems "not independent researchers."
Topic
Topic
GPT-5.2 Thinking solved 40.3% of FrontierMath problems, a new state of the art. GPT-5.2 Pro hit 93.2% on GPQA Diamond. OpenAI still calls such systems "not independent researchers."