{"rewrite":{"id":"r_88b68ed8593ecca407a2d6e0","clusterId":"c_35ea2998d786afa612fd8665","slug":"moonshot-ai-releases-kimi-k3-open-weight-model-with-2-8-trillion-parameters","model":"deepseek-v4-flash:free","headline":"Moonshot AI Releases Kimi K3 Open-Weight Model With 2.8 Trillion Parameters","summary":"Chinese AI company Moonshot AI released the open-weight version of its Kimi K3 model on Hugging Face on July 27, 2026, fulfilling a promise made when the API version launched on July 16. Kimi K3 is a Mixture of Experts model with 2.8 trillion total parameters, activating only 16 of its 896 experts per token. It supports a 1 million token context window and native image input, making it a multimodal model. The released weights are approximately 1.5 terabytes in size, and community quantized GGUF versions appeared shortly after. Running the model locally requires massive GPU memory, with estimates suggesting a server equipped with 12 to 16 NVIDIA H200 GPUs, costing 100 million to 200 million yen, is needed for practical speeds. The release drew significant reactions on X and Reddit, with some calling it a symbol of China's open-weight strategy, while others noted that local execution is impractical and API use is more realistic. Third-party platforms including LM Studio and Ollama have begun hosting Kimi K3, and the technical report is available on GitHub.","whyItMatters":"The open release shifts Kimi K3 from a server-bound API to a model that third-party clouds and research institutions can run and modify, which could accelerate development but also raises questions about misuse risks like cybersecurity.","webCardHtml":"\u003cp\u003eMoonshot AI\u0026#39;s Kimi K3 release follows a pattern of Chinese AI firms open-sourcing large models, but the scale here is notable. The model\u0026#39;s 2.8 trillion parameters and 1.5 terabyte weight size put it beyond the reach of typical consumer hardware, a point the company\u0026#39;s own documentation acknowledges. The sparse MoE design, with only 16 of 896 experts active per token, is what makes the model\u0026#39;s computational demands lower than its total parameter count suggests, though memory requirements remain steep.\u003c/p\u003e\u003cp\u003eReaction on social media has been split. Some users on X and Reddit frame the release as a milestone for open-weight AI from China, while others point out that the size makes local execution unrealistic, steering most users toward API access. The model\u0026#39;s popularity has already strained Moonshot AI\u0026#39;s own servers, leading to a temporary suspension of new subscriptions, and third-party hosting through LM Studio and Ollama is now available as an alternative. The technical report, published on GitHub, accompanies the weights, giving researchers a basis for further work. The estimated hardware cost of 100 million to 200 million yen for a practical local setup means the model is effectively aimed at enterprises and research labs rather than individual users.\u003c/p\u003e","blueskyPost":"Kimi K3's open weights are 1.5 TB and need 12-16 H200 GPUs. Moonshot AI's release is a statement about China's open-weight strategy, not a practical local deployment option.","twitterPost":"Kimi K3's open weights are 1.5 TB, needing 12-16 H200 GPUs. The release is a statement, not a practical local option.","threadsPost":"Kimi K3's open weights are 1.5 TB, and running them at speed needs 12-16 H200 GPUs, costing up to 200 million yen. Moonshot AI's release is a statement about China's open-weight strategy, not a practical local deployment option.","newsletterBlurb":"Moonshot AI has released the open-weight version of its 2.8 trillion parameter Kimi K3 model, making it available on Hugging Face. The model's size limits local use to high-end servers, but third-party cloud hosting is already expanding access.","attributionJson":"[{\"source\":\"ASCII.jp\",\"url\":\"https://ascii.jp/elem/000/004/422/4422583/?rss\",\"title\":\"2.8兆パラメータAI「Kimi K3」公開　動かすには1億円超級サーバが必要？\"},{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20260728-kimi-k3-open/\",\"title\":\"China's High-Performance AI 'Kimi-K3' Open-Sourced as Scheduled, 2.8 Trillion Parameters Approaching Claude Fable 5 and GPT-5.6 Sol Performance\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":7739,"outputTokens":1072,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1785879927,"createdAt":"2026-08-04T21:37:52.000Z","publishedAt":"2026-08-04T21:42:23.000Z","updatedAt":"2026-08-04T21:37:52.000Z"},"cluster":{"id":"c_35ea2998d786afa612fd8665","canonicalTitle":"2.8兆パラメータAI「Kimi K3」公開　動かすには1億円超級サーバが必要？","representativeArticleId":"a_dbbd044a2bb4da15257641b2","sourceCount":2,"writtenSourceCount":2,"writeAttempts":0,"isSolo":false,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[],\"studios\":[],\"people\":[],\"type\":\"announcement\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-07-28T01:30:00.000Z","lastSeenAt":"2026-07-28T01:42:00.000Z","updatedAt":"2026-08-04T21:42:24.000Z"},"attribution":[{"source":"ASCII.jp","url":"https://ascii.jp/elem/000/004/422/4422583/?rss","title":"2.8兆パラメータAI「Kimi K3」公開　動かすには1億円超級サーバが必要？"},{"source":"GIGAZINE","url":"https://gigazine.net/news/20260728-kimi-k3-open/","title":"中国製高性能AI「Kimi-K3」が予定通りオープンモデル化される、2.8兆パラメーターでClaude Fable 5やGPT-5.6 Solに迫る性能"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":[],"studios":[],"people":[],"type":"announcement","domain":"other","is_roundup":false},"keyFacts":["Moonshot AI released the open-weight version of Kimi K3 on Hugging Face on July 27, 2026.","Kimi K3 has 2.8 trillion total parameters and uses a Mixture of Experts architecture with 896 experts, activating 16 per token.","The released weights are approximately 1.5 terabytes in size.","Running Kimi K3 at practical speeds requires a server with 12 to 16 NVIDIA H200 GPUs, costing an estimated 100 million to 200 million yen.","Kimi K3 supports a 1 million token context window and native image input."]}