Le laboratoire chinois Moonshot dévoile un modèle de raisonnement à l'état de l'art sur plusieurs benchmarks de référence.

Avec Kimi K2, la Chine affirme sa montée en puissance sur le marché mondial de l’IA, tout en défendant un modèle plus ouvert ...

Chinese AI startup Moonshot releases open-source Kimi K2 model that outperforms OpenAI and Anthropic on coding tasks with ...

China’s Moonshot AI has launched Kimi K2, a trillion-parameter model that beats ChatGPT and GPT-4 in coding and math tasks, ...

Trillion-parameter open model outshines the big boys Chinese AI upstart Moonshot AI has lobbed a serious challenge at OpenAI ...

Alibaba-backed startup Moonshot released late Friday night its Kimi K2 model as a low-cost, open source large language model, ...

China is rapidly emerging as a major power in the AI world. Following DeepSeek’s success, Moonshot AI, a Chinese AI startup, ...

Discover Kimi K2, the open-source AI model with one trillion parameters, redefining innovation and accessibility in ...

China’s Kimi K2 beats GPT‑4 in internal benchmarks. It excels in reasoning, coding, and creative tasks. But experts say ...

Moonshot AI's Kimi K2, an open-source model, surpasses GPT-4 in benchmarks. Available for free, it offers advanced agentic ...

Moonshot released the Kimi K2 AI model, a new open-source large language model that is positioned as a direct competitor to ...

Moonshot’s Kimi K2 AI model beats GPT-4.1 and Claude in coding benchmarks, offers open-source access, and slashes costs ...