PANews June 11 news, according to the Securities Times, OpenAI announced the launch of a new generation of reasoning model o3-pro, and said that this is the most capable AI model currently. OpenAI said that o3-pro can access tools, which enables it to search the network, analyze files, reason about visual input, use Python, use memory to personalize its responses, etc. According to OpenAI's internal testing, o3-pro surpassed Google Gemini 2.5 Pro in the mathematics benchmark test AIME 2024, and defeated Anthropic Claude 4 Opus in the doctoral-level scientific test GPQA Diamond, showing leading performance in the field of reasoning models. In addition, OpenAI CEO Sam Altman announced on social media that OpenAI's reasoning model o3 has been reduced by 80%.