As the first generation thinking agent with native support for simultaneous thinking and tool usage, kimi k2 thinking achieves perfect fusion of deep thinking and tool orchestration, reaching sota levels in multiple benchmarks and marking a major breakthrough for open source ai reasoning models. Kimi k2 thinking uses test. Moonshot released its new kimi k2 thinking model on thursday Kimi k2 thinking sets ‘new records across benchmarks that assess reasoning, coding and agent capabilities’, moonshot ai researchers say.
WATCH
