按年被提到
20252026 · 2 次2026
强化学习与智能体›大模型智能体·arXiv:2507.20534 · cs.LG

Kimi K2: Open Agentic Intelligence

Kimi Team, Yifan Bai, Yiping Bao, Y. Charles, Cheng Chen, Guanduo Chen, Haiting Chen, Huarong Chen, Jiahao Chen, Ningxin Chen, Ruijue Chen, Yanru Chen 等 200 人 · 2025
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-02-28 · 最近 2026-07-22
2不同的人
2活跃月份
2讨论串
—被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We introduce Kimi K2, a Mixture-of-Experts (MoE) large language model with 32 billion activated parameters and 1 trillion total parameters. We propose the MuonClip optimizer, which improves upon Muon with a novel QK-clip technique to address training instability while enjoying the advanced token efficiency of Muon. Based on MuonClip, K2 was pre-trained on 15.5 trillion tokens with zero loss spike. During post-training, K2 undergoes a multi-stage post-training process, highlighted by a large-scale agentic data synthesis pipeline and a joint reinforcement learning (RL) stage, where the model improves its capabilities through interactions with real and synthetic environments. Kimi K2 achieves state-of-the-art performance among open-source non-thinking models, with strengths in agentic capabilities. Notably, K2 obtains 66.1 on Tau2-Bench, 76.5 on ACEBench (En), 65.8 on SWE-Bench Verified, and 47.3 on SWE-Bench Multilingual -- surpassing most open and closed-sourced baselines in non-thinking settings. It also exhibits strong capabilities in coding, mathematics, and reasoning tasks, with a score of 53.7 on LiveCodeBench v6, 49.5 on AIME 2025, 75.1 on GPQA-Diamond, and 27.1 on OJBench, all without extended thinking. These results position Kimi K2 as one of the most capable open-source large language models to date, particularly in software engineering and agentic tasks. We release our base and post-trained model checkpoints to facilitate future research and applications of agentic intelligence.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论linkregister · 2026-07

“Commenters are overlooking the significance of this information and posting emotional reactions based on perceptions of fairness or feelings of schadenfreude. The economic viability of Anthropic and OpenAI rely on their being able to charge more for model access than their R&D and inference costs. If the market price for SOTA model access drops below tha…”

"We have information that Moonshot distilled Fable for the development of K3"原评论 ↗
被提交为 story · 1 次