[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-how-to-use-claude-4-8-models-in-python-zh":3,"article-related-how-to-use-claude-4-8-models-in-python-zh":30,"series-ai-agent-a708dcdf-cae4-4483-a256-5df230e66543":82},{"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},"a708dcdf-cae4-4483-a256-5df230e66543","how-to-use-claude-4-8-models-in-python-zh","怎麼用 Python 呼叫 Claude 4.8","\u003Cp data-speakable=\"summary\">這篇教你用 Python 和 \u003Ca href=\"\u002Ftag\u002Fanthropic\">Anthropic\u003C\u002Fa> SDK 呼叫 \u003Ca href=\"\u002Ftag\u002Fclaude\">Claude\u003C\u002Fa> 4.8，並加入快取重用重複請求。\u003C\u002Fp>\u003Cp>這篇給已經會寫 Python、想把 Claude 4.8 接進應用程式的開發者。照做完，你會拿到一個可執行的 Python 範例，能發送提示詞、接收模型回覆，還能把重複請求改成快取命中。\u003C\u002Fp>\u003Cp>整\u003Ca href=\"\u002Fnews\u002Faws-explains-ai-agents-workflow-zh\">個流程\u003C\u002Fa>只用官方 Anthropic Python SDK，適合先做聊天助理、客服\u003Ca href=\"\u002Fnews\u002F5-midjourney-alternatives-for-image-creators-zh\">工具\u003C\u002Fa>或內部知識問答的最小可行版本。\u003C\u002Fp>\u003Ch2>開始之前\u003C\u002Fh2>\u003Cul>\u003Cli>Python 3.10+\u003C\u002Fli>\u003Cli>Anthropic 帳號與 API key\u003C\u002Fli>\u003Cli>可連網的終端機\u003C\u002Fli>\u003Cli>pip 23+\u003C\u002Fli>\u003Cli>官方文件：https:\u002F\u002Fdocs.anthropic.com\u002F\u003C\u002Fli>\u003Cli>官方 SDK repo：https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fanthropic-sdk-python\u003C\u002Fli>\u003C\u002Ful>\u003Cp>先把 \u003Ca href=\"\u002Ftag\u002Fapi\">API\u003C\u002Fa> key 放進環境變數或秘密管理工具，不要直接寫進原始碼，這樣之後搬到伺服器也能沿用同一套設定。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780009366539-s0pd.png\" alt=\"怎麼用 Python 呼叫 Claude 4.8\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Ch2>Step 1: 安裝 Anthropic SDK\u003C\u002Fh2>\u003Cp>這一步的目的，是先讓你的 Python 環境具備呼叫 Claude API 的能力。\u003C\u002Fp>\u003Cpre>\u003Ccode>pip install anthropic\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>安裝完成後，進入 Python 互動環境測試匯入。你應該看不到 import error，代表套件已經可以使用。\u003C\u002Fp>\u003Ch2>Step 2: 設定 API key 環境變數\u003C\u002Fh2>\u003Cp>這一步的目的，是讓程式能驗證身分，但不用把密鑰寫死在程式裡。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780009365664-n0as.png\" alt=\"怎麼用 Python 呼叫 Claude 4.8\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cpre>\u003Ccode>export ANTHROPIC_API_KEY=\"YOUR_API_KEY\"\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>如果你在 Windows PowerShell，改用 \u003Ccode>$env:ANTHROPIC_API_KEY=\"YOUR_API_KEY\"\u003C\u002Fcode>。你應該能在 shell 中印出這個變數，並確認內容和 Anthropic 控制台裡的 key 一致。\u003C\u002Fp>\u003Ch2>Step 3: 建立最小 Claude Client\u003C\u002Fh2>\u003Cp>這一步的目的，是先做出一個能送出單一提示詞、並印出單一回覆的 Python 腳本。\u003C\u002Fp>\u003Cpre>\u003Ccode>from anthropic import Anthropic\u003C\u002Fcode>\u003C\u002Fpre>\u003Cpre>\u003Ccode>client = Anthropic(api_key=\"YOUR_API_KEY\"ाएर)\u003C\u002Fcode>\u003C\u002Fpre>\u003Cpre>\u003Ccode>response = client.messages.create(\n    model=\"claude-sonnet-4-8\",\n    max_tokens=1024,\n    messages=[\n        {\"role\": \"user\", \"content\": \"Explain Claude 4.8 in one paragraph.\"}\n    ]\n)\n\nprint(response.content[0].text)\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>執行後，你應該看到\u003Ca href=\"\u002Fnews\u002Fxai-launches-grok-build-terminal-coding-zh\">終端機\u003C\u002Fa>輸出一段模型生成文字，而不是 API 錯誤。這代表模型呼叫已經從程式一路通到服務端。\u003C\u002Fp>\u003Ch2>Step 4: 加入提示詞快取層\u003C\u002Fh2>\u003Cp>這一步的目的，是讓相同提示詞第二次進來時直接回傳快取，減少重複請求。\u003C\u002Fp>\u003Cpre>\u003Ccode>from anthropic import Anthropic\n\nclass OptimizedClient:\n    def __init__(self):\n        self.client = Anthropic(api_key=\"YOUR_API_KEY\")\n        self.cache = {}\n\n    def get_response(self, prompt: str, model: str = \"claude-sonnet-4-8\") -> str:\n        if prompt in self.cache:\n            return self.cache[prompt]\n\n        response = self.client.messages.create(\n            model=model,\n            max_tokens=1024,\n            messages=[{\"role\": \"user\", \"content\": prompt}]\n        )\n        self.cache[prompt] = response.content[0].text\n        return self.cache[prompt]\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>用同一個提示詞連續呼叫兩次。你應該看到兩次輸出內容相同，而且第二次不再需要打 API，而是直接從記憶體快取回傳。\u003C\u002Fp>\u003Ch2>Step 5: 統一模型與 token 設定\u003C\u002Fh2>\u003Cp>這一步的目的，是把模型名稱和輸出長度集中管理，方便你之後切換 Claude 變體。\u003C\u002Fp>\u003Cp>把 model 名稱放在一個地方，避免每個呼叫點都手動改字串。把 \u003Ccode>max_tokens\u003C\u002Fcode> 調成符合任務長度的值，短摘要用較小數字，長分析再提高。\u003C\u002Fp>\u003Cp>接著送一個短提示詞和一個長提示詞。你應該看到短輸入得到精簡回覆，長輸入得到較完整回覆，而且程式只需要改參數，不需要改主流程。\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>重複提示詞處理\u003C\u002Ftd>\u003Ctd>每次都呼叫 API\u003C\u002Ftd>\u003Ctd>相同提示詞直接命中快取\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>程式複雜度\u003C\u002Ftd>\u003Ctd>單一直接呼叫\u003C\u002Ftd>\u003Ctd>可重用 wrapper，含快取與模型參數\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>重複請求成本\u003C\u002Ftd>\u003Ctd>重送一次就再耗一次 token\u003C\u002Ftd>\u003Ctd>第二次請求略過 API\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>常見錯誤\u003C\u002Fh2>\u003Cul>\u003Cli>把 API key 寫死在程式裡。修法：改用 \u003Ccode>ANTHROPIC_API_KEY\u003C\u002Fcode> 環境變數或秘密管理工具。\u003C\u002Fli>\u003Cli>模型名稱拼錯。修法：送出前先對照 Anthropic 文件確認 Claude 4.8 的正確 model 字串。\u003C\u002Fli>\u003Cli>以為相似提示詞也會命中快取。修法：先用完全相同的 prompt 當 key，若需要模糊比對，再加正規化規則。\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>接下來可以看什麼\u003C\u002Fh2>\u003Cp>下一步可以把這個範例升級成可重試、可記錄結構化 log 的版本，再接 Redis 這類持久化快取，讓 Claude 整合在重啟後也能保留命中資料。\u003C\u002Fp>","這篇教你用 Python 和 Anthropic SDK 呼叫 Claude 4.8，並加入快取重用重複請求。","zhuanlan.zhihu.com","https:\u002F\u002Fzhuanlan.zhihu.com\u002Fp\u002F2042174702605657674",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780009366539-s0pd.png","ai-agent","zh","b8096fd3-2102-4649-a1a2-ea0536bbff31",[17,18,19,20,21],"Python","Anthropic SDK","Claude 4.8","API key","cache",[23,24,25],"先安裝 Anthropic SDK，再用環境變數保存 API key。","用最小 client 驗證 Claude 4.8 呼叫，再加上提示詞快取。","把 model 與 max_tokens 集中管理，方便切換與調整輸出長度。",7,"2026-05-28T23:02:20.794444+00:00","2026-05-28T23:02:20.772+00:00","e3b68196-9e64-4c18-a3b6-a73e73bfb367",{"tags":31,"relatedLang":41,"relatedPosts":45},[32,34,36,37,39],{"name":20,"slug":33},"api-key",{"name":17,"slug":35},"python",{"name":21,"slug":21},{"name":18,"slug":38},"anthropic-sdk",{"name":19,"slug":40},"claude-48",{"id":15,"slug":42,"title":43,"language":44},"how-to-use-claude-4-8-models-in-python-en","How to Use Claude 4.8 Models in Python","en",[46,52,58,64,70,76],{"id":47,"slug":48,"title":49,"cover_image":50,"image_url":50,"created_at":51,"category":13},"45c6e678-8ac7-4881-8096-34703d7db136","yong-langgraph-zuo-chu-dai-li-shi-rag-xi-tong-zh","用 LangGraph 做出代理式 RAG 系統","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781485382723-k7xk.png","2026-06-15T01:02:29.343467+00:00",{"id":53,"slug":54,"title":55,"cover_image":56,"image_url":56,"created_at":57,"category":13},"5478bdd3-1241-4185-858f-345b365b24a8","manus-ai-proves-agents-are-ready-for-real-work-zh","Manus AI 證明代理人已能上工，但定價會決定它的命運","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781444874515-g9dg.png","2026-06-14T13:47:21.276926+00:00",{"id":59,"slug":60,"title":61,"cover_image":62,"image_url":62,"created_at":63,"category":13},"7ea0ef5b-d12c-4b18-b8fd-6ae3de67c296","coinbase-ai-agent-accounts-strict-limits-zh","Coinbase 讓 AI 代理代交易與代支付是對的，但前提是嚴格限權","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781409758550-mjql.png","2026-06-14T04:02:15.334232+00:00",{"id":65,"slug":66,"title":67,"cover_image":68,"image_url":68,"created_at":69,"category":13},"7315dc1e-d3c0-4888-8466-1328e8819be0","peft-llm-fine-tuning-without-full-retraining-zh","PEFT LoRA 微調 LLM 實作指南","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781403475967-xlpz.png","2026-06-14T02:17:26.268208+00:00",{"id":71,"slug":72,"title":73,"cover_image":74,"image_url":74,"created_at":75,"category":13},"5e2ed9f7-4240-429b-97c7-ffd31e4a45ee","llm-research-engineers-post-training-services-zh","LLM研究工程師把後訓練做成服務","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781402598646-2jzs.png","2026-06-14T02:02:46.765352+00:00",{"id":77,"slug":78,"title":79,"cover_image":80,"image_url":80,"created_at":81,"category":13},"09e34016-bbc0-4313-b090-2dbfdd6cf96a","fine-tuning-slms-turns-enterprise-ai-practical-zh","SLM 微調把企業 AI 變可用","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781359406320-5jrq.png","2026-06-13T14:02:55.242488+00:00",[83,88,93,98,103,108,113,118,123,128],{"id":84,"slug":85,"title":86,"created_at":87},"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":89,"slug":90,"title":91,"created_at":92},"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":94,"slug":95,"title":96,"created_at":97},"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":99,"slug":100,"title":101,"created_at":102},"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":104,"slug":105,"title":106,"created_at":107},"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":109,"slug":110,"title":111,"created_at":112},"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":114,"slug":115,"title":116,"created_at":117},"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":119,"slug":120,"title":121,"created_at":122},"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":124,"slug":125,"title":126,"created_at":127},"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":129,"slug":130,"title":131,"created_at":132},"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"]