OpenAI’s Gemini Deep Think and OpenAI’s AI reasoning model both earned gold medals at the International Mathematical Olympiad (IMO), scoring 35 out of 42. They tackled five out of six problems successfully, showcasing the power of general-purpose AI reasoning models that efficiently utilize natural language input. Meanwhile, ByteDance’s Seed-Prover achieved a Silver rating by solving four problems. Alibaba’s Qwen team launched cutting-edge open-source generative AI models, enhancing reasoning and coding capabilities. Their Qwen3 models show significant advancements over competitors, offering up to a 256K-token context and achieving top benchmark scores. Google introduced new tools such as Opal for application creation and updated AI features for enhanced search and virtual try-ons. Nvidia updated the Nemotron models, while Anthropic and Sapient Intelligence revealed robust new architectures for advanced reasoning. The recent White House “AI Action Plan” aims to accelerate AI development and innovation in the U.S., receiving positive feedback from the tech community.
Source link
Share
Read more