Benchmarks

A chart by METR, a nonprofit AI organization, has become an industrywide obsession as it tracks the rapid development of large AI systems.

Alibaba confirmed it secretly developed HappyHorse-1.0, an AI video model that debuted at the top of global benchmarks, surpassing rivals with audio-visual synchronization capabilities.

A new benchmark called APEX-Agents shows that even leading AI models like GPT-5.2 and Gemini 3 Flash fail on most complex, multi-domain tasks drawn from professional fields like law and finance, raising doubts about their immediate readiness for the workplace.