"On Thursday, the Beijing-based company released Kimi K2 Thinking, a reasoning model that it says outperforms OpenAI's GPT-5 and Anthropic's Claude Sonnet 4.5 on key benchmarks, including Humanity's Last Exam, BrowseComp (which tests AI agents' ability to extract hard-to-find online information via web browsers), and Seal-0 (which assesses reasoning capabilities). Kimi K2 Thinking also showed coding abilities that were comparable to GPT-5 and Sonnet 4.5, but not notably more impressive."
"Kimi K2 Thinking is a Mixture-of-Experts (MoE) model blending long-horizon planning, adaptive reasoning, and the use of online tools (such as browsers), "continually generating and refining hypotheses, verifying evidence, reasoning, and constructing coherent answers," the company wrote. "This interleaved reasoning allows it to decompose ambiguous, open-ended problems into clear, actionable subtasks." It was trained with around 1 trillion parameters and can be accessed on Hugging Face."
Moonshot released Kimi K2 Thinking, an open-source reasoning model built as a Mixture-of-Experts combining long-horizon planning, adaptive reasoning, and online tool use. The model interleaves hypothesis generation, evidence verification, and iterative reasoning to decompose ambiguous, open-ended problems into actionable subtasks. Kimi K2 Thinking was trained with around one trillion parameters and is available on Hugging Face. Benchmark results claim K2 outperforms GPT-5 and Claude Sonnet 4.5 on Humanity's Last Exam, BrowseComp, and Seal-0 while showing coding abilities comparable to those models. The open-source release intensifies competition between Chinese labs and proprietary US AI providers.
Read at ZDNET
Unable to calculate read time
Collection
[
|
...
]