[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-kimi-k2-5-local-setup-ollama-docker-zh":3,"article-related-kimi-k2-5-local-setup-ollama-docker-zh":30,"series-ai-agent-be07f530-b13a-4c07-ada2-f93f112970e3":75},{"id":4,"slug":5,"title":6,"content":7,"summary":8,"source":9,"source_url":10,"author":11,"image_url":12,"cover_image":12,"category":13,"language":14,"translated_content":11,"related_article_id":15,"keywords":16,"key_takeaways":22,"views":26,"created_at":27,"published_at":28,"topic_cluster_id":29},"be07f530-b13a-4c07-ada2-f93f112970e3","kimi-k2-5-local-setup-ollama-docker-zh","Kimi-K2.5 本機部署產出","\u003Cp data-speakable=\"summary\">用 \u003Ca href=\"\u002Ftag\u002Fdocker\">Docker\u003C\u002Fa> 和 Ollama 在本機部署 Kimi-K2.5，完成離線推理與可驗證的服務啟動。\u003C\u002Fp>\u003Cp>這篇給想把 Kimi-K2.5 跑在自己電腦上的開發者看，不需要先做雲端架構或額外\u003Ca href=\"\u002Fnews\u002F2026-fangdiancai-daibi-hua-zui-qiang-5-pingtai-zh\">平台\u003C\u002Fa>整合。照著做完，你會得到一組可重複啟動的 Docker Compose 服務、一個已下載的模型，以及一個能直接回應提示詞的本機\u003Ca href=\"\u002Fnews\u002Fqodo-2-8-multi-repo-ai-code-review-beta-zh\">測試\u003C\u002Fa>結果。\u003C\u002Fp>\u003Ch2>開始之前\u003C\u002Fh2>\u003Cul>\u003Cli>Docker Desktop 4.30+ 或 Docker Engine 24+，並啟用 Docker Compose v2\u003C\u002Fli>\u003Cli>本機已安裝 Ollama，或可直接拉取 \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Follama\u002Follama\">Ollama GitHub\u003C\u002Fa> 的容器映像\u003C\u002Fli>\u003Cli>Node 20+，僅在你要用腳本測試 API 時需要\u003C\u002Fli>\u003Cli>至少 16 GB RAM，較大的量化模型建議更多\u003C\u002Fli>\u003Cli>至少 50 GB 可用磁碟空間，放模型檔、快取與後續更新\u003C\u002Fli>\u003Cli>GitHub 帳號，若你要版本控管 compose 檔\u003C\u002Fli>\u003Cli>可查閱 \u003Ca href=\"https:\u002F\u002Follama.com\u002Fdocs\">Ollama docs\u003C\u002Fa> 以確認模型指令與映像名稱\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>Step 1: 建立專案資料夾\u003C\u002Fh2>\u003Cp>先建立一個乾淨工作目錄，讓 compose 檔、模型\u003Ca href=\"\u002Fnews\u002Fai-demand-starts-paying-for-data-centers-zh\">資料\u003C\u002Fa>與記錄檔分開管理，之後調整設定也不會混亂。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782828169969-prdn.png\" alt=\"Kimi-K2.5 本機部署產出\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cpre>\u003Ccode>mkdir kimi-k2-5-local\ncd kimi-k2-5-local\nmkdir models data logs\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>你應該看到三個新資料夾已建立，且終端機目前仍停在專案根目錄。\u003C\u002Fp>\u003Ch2>Step 2: 寫入 Compose 設定檔\u003C\u002Fh2>\u003Cp>接著定義一份可重複啟動的容器設定，讓 Ollama 服務固定對外開放本機 \u003Ca href=\"\u002Ftag\u002Fapi\">API\u003C\u002Fa>。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782828171451-hsy3.png\" alt=\"Kimi-K2.5 本機部署產出\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cpre>\u003Ccode>services:\n  ollama:\n    image: ollama\u002Follama:latest\n    container_name: kimi-k2-5-ollama\n    ports:\n      - \"11434:11434\"\n    volumes:\n      - .\u002Fmodels:\u002Froot\u002F.ollama\n    restart: unless-stopped\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>你應該看到根目錄出現 \u003Ccode>docker-compose.yml\u003C\u002Fcode>，而且服務名稱在 Docker 裡可清楚辨識。\u003C\u002Fp>\u003Ch2>Step 3: 啟動 Ollama 容器\u003C\u002Fh2>\u003Cp>現在把 runtime 跑起來，讓本機先具備 Ollama 服務，再進行模型下載與推理測試。\u003C\u002Fp>\u003Cpre>\u003Ccode>docker compose up -d\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>你應該看到容器成功啟動，並且 \u003Ccode>docker ps\u003C\u002Fcode> 會列出 \u003Ccode>kimi-k2-5-ollama\u003C\u002Fcode> 正在執行。\u003C\u002Fp>\u003Ch2>Step 4: 下載 Kimi-K2.5 模型\u003C\u002Fh2>\u003Cp>服務啟動後，拉取你要使用的模型標籤，讓本機先備妥權重檔，之後才能直接做推理。\u003C\u002Fp>\u003Cpre>\u003Ccode>docker exec -it kimi-k2-5-ollama ollama pull kimi-k2.5\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>你應該看到下載完成，沒有 layer 失敗訊息，而且模型會出現在 Ollama 的模型清單中。\u003C\u002Fp>\u003Ch2>Step 5: 執行本機提示詞測試\u003C\u002Fh2>\u003Cp>最後送出一個簡單提示詞，確認模型能透過本機 Ollama 端點正常回應。\u003C\u002Fp>\u003Cpre>\u003Ccode>docker exec -it kimi-k2-5-ollama ollama run kimi-k2.5 \"Write one sentence about local AI development.\"\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>你應該看到終端機輸出一段文字回覆，這代表模型已經能在你的電腦上處理請求。\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>指標\u003C\u002Fth>\u003Cth>基準／優化前\u003C\u002Fth>\u003Cth>結果／優化後\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>部署方式\u003C\u002Ftd>\u003Ctd>手動安裝與逐步設定\u003C\u002Ftd>\u003Ctd>Docker Compose 一鍵啟動\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>記憶體規劃\u003C\u002Ftd>\u003Ctd>沒有明確門檻\u003C\u002Ftd>\u003Ctd>建議至少 16 GB RAM\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>磁碟規劃\u003C\u002Ftd>\u003Ctd>臨時估算空間\u003C\u002Ftd>\u003Ctd>建議預留 50 GB 可用空間\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>常見錯誤\u003C\u002Fh2>\u003Cul>\u003Cli>記憶體不足：如果容器退出或頻繁交換記憶體，請增加 RAM，或改用更小的量化模型標籤。\u003C\u002Fli>\u003Cli>忘記開放 11434：如果 API 無法連線，先確認 compose 檔有正確對映埠號，且沒有被其他服務占用。\u003C\u002Fli>\u003Cli>模型名稱打錯：如果 Ollama 顯示找不到模型，請回頭核對標籤拼字，並使用文件或 repo 中的完整名稱。\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>接下來可以看什麼\u003C\u002Fh2>\u003Cp>本機部署成功後，可以再接一個客戶端應用、量測提示詞延遲，或把服務放到反向代理後面，讓同一個局域網內的其他裝置也能安全使用。\u003C\u002Fp>","用 Docker 和 Ollama 在本機部署 Kimi-K2.5，完成離線推理與可驗證的服務啟動。","okanmorkoc.com","https:\u002F\u002Fokanmorkoc.com\u002Fhow-to-setup-kimi-k2-5-locally-via-ollama-2-easy-build\u002F",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782828169969-prdn.png","ai-agent","zh","a72fc4a2-7e7d-4f06-b34a-857d65ad30e2",[17,18,19,20,21],"Docker","Ollama","Kimi-K2.5","Docker Compose","本機部署",[23,24,25],"用 Docker Compose 可以把 Kimi-K2.5 的本機部署流程固定化。","先確認記憶體與磁碟空間，再下載模型與測試推理。","成功標準是容器在跑、模型已下載、終端機能回應提示詞。",0,"2026-06-30T14:02:22.214135+00:00","2026-06-30T14:02:22.18+00:00","e3b68196-9e64-4c18-a3b6-a73e73bfb367",{"tags":31,"relatedLang":34,"relatedPosts":38},[32],{"name":17,"slug":33},"docker",{"id":15,"slug":35,"title":36,"language":37},"kimi-k2-5-local-setup-ollama-docker-en","Kimi-K2.5 Local Setup with Ollama and Docker","en",[39,45,51,57,63,69],{"id":40,"slug":41,"title":42,"cover_image":43,"image_url":43,"created_at":44,"category":13},"5dea881b-6fa6-4193-a0e7-3e0d391ae785","happycapy-best-manus-alternative-zh","HappyCapy 才是 Manus 最佳替代品","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782831781342-7506.png","2026-06-30T15:02:34.343828+00:00",{"id":46,"slug":47,"title":48,"cover_image":49,"image_url":49,"created_at":50,"category":13},"1eca3b84-685f-4695-ae9b-3d16676de034","cursor-ai-code-review-fading-zh","Cursor 數據顯示 AI 審查在退位","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782820977595-qxyn.png","2026-06-30T12:02:30.802192+00:00",{"id":52,"slug":53,"title":54,"cover_image":55,"image_url":55,"created_at":56,"category":13},"cd51c43c-312b-4bcf-a6b2-b3217c4e05b7","llm-wikis-beat-raw-rag-knowledge-work-zh","LLM 維護的 wiki 比原始 RAG 更適合真正的知識工作","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782760669415-7e3l.png","2026-06-29T19:17:20.761542+00:00",{"id":58,"slug":59,"title":60,"cover_image":61,"image_url":61,"created_at":62,"category":13},"cde225a8-eb8e-4724-a089-77f36af0e8a6","mcps-new-primitives-make-agent-middleware-obsolete-zh","MCP 的新原語，正在淘汰自製 agent middleware","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782748974384-f5w9.png","2026-06-29T16:02:24.789168+00:00",{"id":64,"slug":65,"title":66,"cover_image":67,"image_url":67,"created_at":68,"category":13},"6e37d84c-aa27-4d4d-bbf1-81c47dc4522d","mcp-servers-ai-workflows-explained-zh","MCP Server 讓 AI 工具接上工作流","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782747180723-q3gs.png","2026-06-29T15:32:33.536175+00:00",{"id":70,"slug":71,"title":72,"cover_image":73,"image_url":73,"created_at":74,"category":13},"a5333ae2-bfd1-434a-92dd-575e824538c3","openmontage-open-source-ai-video-production-zh","OpenMontage 證明 AI 影片製作該由開源接管","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782685072512-v02f.png","2026-06-28T22:17:22.846394+00:00",[76,81,86,91,96,101,106,111,116,121],{"id":77,"slug":78,"title":79,"created_at":80},"4ae1e197-1d3d-4233-8733-eafe9cb6438b","claude-now-uses-your-pc-to-finish-tasks-zh","Claude 開始幫你操作電腦","2026-03-26T07:20:48.457387+00:00",{"id":82,"slug":83,"title":84,"created_at":85},"5bede67f-e21c-413d-9ab8-54a3c3d26227","googles-2026-ai-agent-report-decoded-zh","Google 2026 AI Agent 報告解讀","2026-03-26T11:15:22.651956+00:00",{"id":87,"slug":88,"title":89,"created_at":90},"2987d097-563f-46c7-b76f-b558d8ef7c2b","kimi-k25-review-stronger-still-not-legend-zh","Kimi K2.5 評測：更強，但還不是神作","2026-03-27T07:15:55.277513+00:00",{"id":92,"slug":93,"title":94,"created_at":95},"95c9053b-e3f4-4cb5-aace-5c54f4c9e044","claude-code-controls-mac-desktop-zh","Claude Code 也能操控 Mac 了","2026-03-28T03:01:58.58121+00:00",{"id":97,"slug":98,"title":99,"created_at":100},"dc58e153-e3a8-4c06-9b96-1aa64eabbf5f","cloudflare-100x-faster-ai-agent-sandbox-zh","Cloudflare 的 AI 沙箱跑超快","2026-03-28T03:09:44.142236+00:00",{"id":102,"slug":103,"title":104,"created_at":105},"1c8afc56-253f-47a2-979f-1065ff072f2a","openai-backs-isara-agent-swarm-bet-zh","OpenAI 挺 Isara 的 agent swarm …","2026-03-28T03:15:27.513155+00:00",{"id":107,"slug":108,"title":109,"created_at":110},"7379b422-576e-45df-ad5a-d57a0d9dd467","openai-plan-automated-ai-researcher-zh","OpenAI 想做自動化 AI 研究員","2026-03-28T03:17:42.090548+00:00",{"id":112,"slug":113,"title":114,"created_at":115},"48c9889e-86df-450b-a356-e4a4b7c83c5b","harness-engineering-ai-agent-reliability-2026-zh","駕馭工程：從「馬具」到「作業系統」，AI Agent 可靠性的終極密碼","2026-03-31T06:42:53.556721+00:00",{"id":117,"slug":118,"title":119,"created_at":120},"96d8e8c8-1edd-475d-9145-b1e7a1b02b65","mcp-explained-from-prompts-to-production-zh","MCP 怎麼把提示詞變工作流","2026-04-01T09:24:39.321274+00:00",{"id":122,"slug":123,"title":124,"created_at":125},"f2ca7720-b471-4ce5-9336-2a9ac2a876fd","amazon-bedrock-agents-multi-agent-workflows-zh","Amazon Bedrock Agents 進入多代理工作流","2026-04-01T09:30:29.945429+00:00"]