archiveAI benchmarks

Chinese leading food delivery and e-commerce service platform Meituan has recently opened its AI model, LongCat-Flash-Thinking™. This model, with 560 billion parameters, aims to challenge OpenAI's GPT-5 in terms of reasoning ability, especially in challenging domains such as formal logic, mathematics, and programming. What Is LongCat-Flash-Thinking? LongCat-Flash-Thinking is a massive...

In the ever-changing field of artificial intelligence, it is crucial to have AI agents that work well in real-world situations. The release of Gaia2, a more advanced and built-in benchmark within the Meta Agents Research Environments (ARE), moves AI agent evaluation beyond what has largely been limited to simple metrics...

In the rapidly developing field of artificial intelligence, China has become the wild card. Kimi-K2 is China’s answer to the DeepSeek model, which caused a stir in the AI space around the world and was a successful demonstration of the biggest open-weight AI project in China. This is not simply...