MOONSHOT AI

About Moonshot AI

Chinese AI company behind Kimi Chat, focusing on long-context language models.

Recent News about Moonshot AI

• China’s AI token economy surges as MiniMax and Zhipu post big gains

  On April 3, 2026 at 14:49 local time in Shanghai, Xinhua reported that Chinese labs MiniMax, Zhipu AI and Moonshot AI are driving a global ‘token economy’ with rapid growth in API usage and overseas adoption. Zhipu AI’s 2025 revenue jumped 131.9% year‑on‑year with token sales up 292.6%, while MiniMax’s 2025 revenue rose 158.9% with about 70% from international markets, and Moonshot’s Kimi K2.5 model was recently adopted as the base engine for U.S. coding platform Cursor.

  Source: People’s Daily Online2026-04-03T06:49:00.000Z

• China’s AGI-Next summit backs 2026 shift from chatbots to agents

  On January 12, 2026 (Beijing time), Chinese tech outlet Huxiu published a detailed account of the AGI-Next summit held at Tsinghua University, where leaders from Zhipu AI, Moonshot AI (Kimi), Alibaba’s Qwen team and Tencent discussed the next phase of China’s AGI ambitions. Speakers argued that 2026 will mark a shift from parameter races and chatbots toward high‑efficiency, task‑oriented AI agents and embodied intelligence.([huxiu.com](https://www.huxiu.com/article/4825438.html))

  Source: Huxiu2026-01-11T16:11:00.000Z

• China’s AI titans race to control next-gen digital gateways

  Caixin reports that Chinese internet giants like Alibaba and ByteDance are shifting from pure model scaling to a multi-front competition over users, hardware, talent and capital in AI. The piece details soaring usage of apps like ByteDance’s Doubao, massive MaaS growth, and IPO plans from key startups such as Zhipu AI and MiniMax.

  Source: Caixin Global2025-12-28T22:31:00.000Z

• Amazon Bedrock adds 18 fully managed open‑weight models from leading AI providers

  Amazon Web Services has expanded Amazon Bedrock with 18 additional fully managed open‑weight models from providers including Mistral AI (Mistral Large 3 and the Ministral 3 series), Google (Gemma 3), Moonshot AI (Kimi K2 Thinking), MiniMax (MiniMax M2), NVIDIA (Nemotron), OpenAI (GPT OSS Safeguard) and Qwen. The move brings Bedrock’s catalog to nearly 100 serverless models, reinforcing AWS’s strategy of offering enterprises a broad, switchable model lineup for agentic workflows, multimodal reasoning, coding assistance and content safety without changing application code.

  Source: PR Newswire Asia (PRNASIA)2025-12-11T02:23:00.000Z

• Nvidia says new AI server boosts Chinese mixture‑of‑experts models tenfold

  Nvidia released benchmark data showing its latest AI server, which packs 72 of its top chips into a single system, can deliver roughly a 10x performance gain when serving large mixture‑of‑experts models such as Moonshot AI’s Kimi K2 Thinking and DeepSeek’s models. The results aim to show that even as some new models train more efficiently, Nvidia’s high‑end servers remain critical for large‑scale inference, reinforcing its dominance against rivals like AMD and Cerebras in the AI deployment market.

  Source: Reuters2025-12-03T17:33:20.000Z

• AI’s safety features can be circumvented with poetry, research finds

  Researchers at Italy’s Icaro Lab showed that large language models from nine major AI companies can be ‘jailbroken’ using carefully crafted poems that embed requests for harmful content. Testing 20 poems against 25 models, they found 62% of prompts elicited unsafe responses, highlighting a serious vulnerability in current safety guardrails and prompting calls for companies to harden defences against such adversarial attacks.

  Source: The Guardian2025-11-30T14:00:00.000Z