[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-ru-he-yong-python-zheng-he-mistral-ocr-zh":3,"article-related-ru-he-yong-python-zheng-he-mistral-ocr-zh":36,"series-tools-3d1c3a08-1490-4e9e-98ef-14079e3056d6":89},{"id":4,"title":5,"content":6,"summary":7,"source":8,"source_url":9,"author":10,"image_url":11,"keywords":12,"language":18,"translated_content":10,"views":19,"is_premium":20,"created_at":21,"updated_at":21,"cover_image":11,"published_at":22,"rewrite_status":23,"rewrite_error":10,"rewritten_from_id":24,"slug":25,"category":26,"related_article_id":27,"status":28,"google_indexed_at":29,"x_posted_at":10,"tweet_text":10,"title_rewritten_at":10,"title_original":10,"key_takeaways":30,"topic_cluster_id":34,"embedding":35,"is_canonical_seed":20},"3d1c3a08-1490-4e9e-98ef-14079e3056d6","如何用 Python 串接 Mistral OCR","\u003Cp data-speakable=\"summary\">這篇教你用 Python 串接 \u003Ca href=\"\u002Fnews\u002Fmistral-cybersecurity-model-banks-europe-zh\">Mist\u003C\u002Fa>ral OCR，完成雲端 PDF、在地檔案與圖片掃描的文字擷取，並把結果保存成可再處理的結構化輸出。\u003C\u002Fp>\u003Cp>這篇給要把 PDF、掃描檔、收據照片轉成可用文字與版面的開發者看。你照著做完，會得到一組可直接執行的 Python 流程，能處理遠端 PDF、上傳本機檔案、保留 Markdown 結構，還能把圖片一併存下來。\u003C\u002Fp>\u003Cp>你也會完成 \u003Ca href=\"\u002Ftag\u002Fapi\">API\u003C\u002Fa> 金鑰設定、驗證輸出結果，並建立一個可延伸到正式環境的 OCR 範例專案。整篇都是照做型步驟，適合第一次接 Mistral OCR 的人。\u003C\u002Fp>\u003Ch2>開始之前\u003C\u002Fh2>\u003Cul>\u003Cli>Python 3.9+\u003C\u002Fli>\u003Cli>pip 23+\u003C\u002Fli>\u003Cli>Mistral AI 帳號，先登入 \u003Ca href=\"https:\u002F\u002Fconsole.mistral.ai\u002F\">La Plateforme\u003C\u002Fa>\u003C\u002Fli>\u003Cli>Mistral API key\u003C\u002Fli>\u003Cli>可連網，因為這是雲端 OCR\u003C\u002Fli>\u003Cli>可測試的 PDF、JPG 或 PNG 檔案\u003C\u002Fli>\u003Cli>Git 版本控制工具，若你要保存範例專案\u003C\u002Fli>\u003C\u002Ful>\u003Cp>先把套件裝在獨立虛擬環境，避免和其他專案的依賴衝突。這樣你之後重跑 OCR 範例時，也比較容易重現相同結果。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1779001446578-9gfa.png\" alt=\"如何用 Python 串接 Mistral OCR\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Ch2>Step 1: 建立 Python 虛擬環境\u003C\u002Fh2>\u003Cp>目的：先做出乾淨的執行環境，讓 Mistral SDK 可以獨立安裝，不會被其他專案套件干擾。\u003C\u002Fp>\u003Cpre>\u003Ccode>python3 -m venv .venv\nsource .venv\u002Fbin\u002Factivate\npip install --upgrade pip\npip install mistralai python-dotenv datauri\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：執行 \u003Ccode>python -c \"import mistra\u003Ca href=\"\u002Fnews\u002Fbitcoin-tops-80k-senate-advances-clarity-act-zh\">la\u003C\u002Fa>i; print('ok')\"\u003C\u002Fcode>，你應該看到 \u003Ccode>ok\u003C\u002Fcode>，代表 SDK 已經安裝成功。\u003C\u002Fp>\u003Ch2>Step 2: 儲存 API 金鑰\u003C\u002Fh2>\u003Cp>目的：把憑證放進環境變數，避免直接寫進程式碼或上傳到版本庫。Mistral OCR 需要先通過驗證，才可以處理文件。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1779001456365-xhog.png\" alt=\"如何用 Python 串接 Mistral OCR\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cpre>\u003Ccode>cat > .env <<'EOF'\nMISTRAL_API_KEY=your_api_key_here\nEOF\n\ncat > .gitignore <<'EOF'\n.env\n.venv\nEOF\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：你應該看到 \u003Ccode>.env\u003C\u002Fcode> 與 \u003Ccode>.\u003Ca href=\"\u002Fnews\u002Fsifive-p570-gen3-rva23-platform-core-zh\">ve\u003C\u002Fa>nv\u003C\u002Fcode> 被 Git 忽略，且在 Mistral 控制台中可確認 API key 已啟用。\u003C\u002Fp>\u003Ch2>Step 3: 初始化 OCR 用戶端\u003C\u002Fh2>\u003Cp>目的：讀取環境變數並建立已驗證的客戶端物件，讓腳本具備呼叫 OCR 端點的能力。\u003C\u002Fp>\u003Cpre>\u003Ccode>from dotenv import load_dotenv\nfrom mistralai import Mistral\nimport os\n\nload_dotenv()\napi_key = os.environ[\"MISTRAL_API_KEY\"]\nclient = Mistral(api_key=api_key)\nprint(\"client-ready\")\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：執行後你應該看到 \u003Ccode>client-ready\u003C\u002Fcode>。如果沒有出現，先檢查 \u003Ccode>.env\u003C\u002Fcode> 是否真的有讀到金鑰。\u003C\u002Fp>\u003Ch2>Step 4: 擷取遠端 PDF 內容\u003C\u002Fh2>\u003Cp>目的：直接對公開 PDF 進行 OCR，快速確認整條流程是否正常，不必先處理檔案上傳。\u003C\u002Fp>\u003Cpre>\u003Ccode>ocr_response = client.ocr.process(\n    model=\"mistral-ocr-latest\",\n    document={\n        \"type\": \"document_url\",\n        \"document_url\": \"https:\u002F\u002Farxiv.org\u002Fpdf\u002F2501.00663\"\n    }\n)\n\nprint(len(ocr_response.pages))\nprint(ocr_response.pages[0].markdown[:800])\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：你應該看到頁數大於 0，並且前幾段 Markdown 會包含標題、段落或圖表引用，而不是一整坨平面文字。\u003C\u002Fp>\u003Ch2>Step 5: 上傳本機檔案並保存圖片\u003C\u002Fh2>\u003Cp>目的：處理存在你電腦上的 PDF，並把文件中的圖片一併保留下來。這一步適合私人文件、內網文件或不能公開上傳的資料。\u003C\u002Fp>\u003Cpre>\u003Ccode>from datauri import parse\n\nuploaded = client.files.upload(\n    file={\"file_name\": \"report.pdf\", \"content\": open(\"report.pdf\", \"rb\")},\n    purpose=\"ocr\"\n)\nsigned_url = client.files.get_signed_url(file_id=uploaded.id)\n\nocr_response = client.ocr.process(\n    model=\"mistral-ocr-latest\",\n    document={\"type\": \"document_url\", \"document_url\": signed_url.url},\n    include_image_base64=True\n)\n\nfor page in ocr_response.pages:\n    for img in page.images:\n        data = parse(img.image_base64)\n        with open(img.id, \"wb\") as f:\n            f.write(data.data)\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：你應該看到一或多個圖片檔寫入磁碟，例如 \u003Ccode>img-0.jpeg\u003C\u002Fcode>。同時 Markdown 仍保有內文與圖像位置，代表版面資訊有被保留。\u003C\u002Fp>\u003Ch2>Step 6: 測試掃描圖並檢查品質\u003C\u002Fh2>\u003Cp>目的：對掃描檔或手機拍照圖做 OCR，確認輸出是否保留標題、表格結構與可讀文字。這一步能幫你判斷哪些文件可直接自動化，哪些還需要清理規則。\u003C\u002Fp>\u003Cpre>\u003Ccode>ocr_response = client.ocr.process(\n    model=\"mistral-ocr-latest\",\n    document={\n        \"type\": \"image_url\",\n        \"image_url\": \"https:\u002F\u002Fexample.com\u002Freceipt.png\"\n    }\n)\n\nprint(ocr_response.pages[0].markdown)\u003C\u002Fcode>\u003C\u002Fpre>\u003Cp>驗收：你應該看到與畫面內容相符的文字，而不是只有零碎字元。若是發票或收據，品項與總金額應該仍然可讀，足以交給後續解析流程。\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>多樣文件準確率\u003C\u002Ftd>\u003Ctd>83.4%，使用 Google Document AI\u003C\u002Ftd>\u003Ctd>約 94.9%，使用 Mistral OCR\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>多樣文件準確率\u003C\u002Ftd>\u003Ctd>89.5%，使用 Azure OCR\u003C\u002Ftd>\u003Ctd>約 94.9%，使用 Mistral OCR\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>吞吐量\u003C\u002Ftd>\u003Ctd>一般單檔人工處理\u003C\u002Ftd>\u003Ctd>單一 GPU 節點最高每分鐘 2,000 頁\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>價格\u003C\u002Ftd>\u003Ctd>團隊人工複核成本浮動\u003C\u002Ftd>\u003Ctd>約每 1,000 頁 1 美元，約每頁 0.001 美元\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>請求限制\u003C\u002Ftd>\u003Ctd>臨時檔案處理\u003C\u002Ftd>\u003Ctd>單次請求最高 50 MB 或 1,000 頁\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>常見錯誤\u003C\u002Fh2>\u003Cul>\u003Cli>API key 遺失或過期。修法：重新檢查 \u003Ccode>.env\u003C\u002Fcode> 裡的 \u003Ccode>MISTRAL_API_KEY\u003C\u002Fcode>，必要時到控制台重新產生金鑰。\u003C\u002Fli>\u003Cli>把本機路徑直接丟進 \u003Ccode>document_url\u003C\u002Fcode>。修法：先上傳檔案，再使用 \u003Ccode>client.files.get_signed_url()\u003C\u002Fcode> 回傳的連結。\u003C\u002Fli>\u003Cli>以為 OCR 只會回傳純文字。修法：改成讀取 Markdown 結構與圖片引用，再決定要不要保留標題、表格與圖示。\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>接下來可以看什麼\u003C\u002Fh2>\u003Cp>當基本 OCR 流程跑通後，下一步可以加入分段切塊、表格欄位抽取與驗證規則，讓文件輸出變成可搜尋知識、可解析發票欄位，或可供檢索的內容。若你要做更大的文件管線，可以再接向量資料庫、Markdown 表格解析器與低信心頁面的人工複核流程。\u003C\u002Fp>","這篇教你用 Python 串接 Mistral OCR，完成雲端 PDF、在地檔案與圖片掃描的文字擷取，並把結果保存成可再處理的結構化輸出。","cohorte.co","https:\u002F\u002Fcohorte.co\u002Fblog\u002Fmistral-ocr-a-deep-dive-into-next-generation-document-understanding",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1779001446578-9gfa.png",[13,14,15,16,17],"Mistral OCR","Python","PDF OCR","Markdown","API 金鑰","zh",2,false,"2026-05-17T07:03:37.885677+00:00","2026-05-17T07:03:37.875+00:00","done","a2f71617-167b-4a6e-86d8-e5e4aea20e8e","ru-he-yong-python-zheng-he-mistral-ocr-zh","tools","b70f16f2-9ed6-4820-8cb6-6b52bca1e7df","published","2026-05-17T09:00:14.126+00:00",[31,32,33],"先用虛擬環境與 .env 建立可重現的 Python OCR 專案。","可用遠端 PDF、上傳檔案與圖片掃描三種方式驗證 Mistral OCR。","輸出不只文字，還包含 Markdown 結構與圖片資訊，適合後續自動化處理。","c3c88dd2-a940-438a-b359-0e5a24562273","[0.006304814,0.0059162877,0.016727408,-0.08679379,-0.0077564917,-0.008213008,-0.016503204,0.002634179,0.0099592265,-0.010801135,-0.014280956,-0.029328959,-0.0008771247,-0.0010131384,0.13037483,-0.0014647479,0.0004765662,0.014714335,0.012939147,-0.026647143,-0.01542793,0.020948851,0.013749007,0.006706052,0.0072271638,-0.0053981133,0.034607656,0.032432333,0.037822682,-0.0031260059,0.024208924,0.002568423,-0.0058785444,0.0050419,-0.01077146,0.0037818167,0.020588504,-0.0093234135,0.005574796,0.023465166,-0.037334226,-0.014813031,0.008023952,0.0012068611,-0.004397398,0.0016560927,-0.00044221253,-0.008088667,0.0039941943,-0.0044755884,0.0041827895,0.014395265,-0.021090321,-0.16574763,0.0049835765,0.009620084,-5.485923e-05,0.008040555,0.041920725,-0.016235117,-0.009462143,0.034937207,0.012915344,0.0056160293,-0.015157915,-0.029774768,0.033433404,0.029044235,-0.011170329,-0.007289478,-0.0024125369,0.024661925,-0.0071985642,-0.012140967,-0.014460829,-0.03920112,0.022309683,-0.014738889,0.01833721,-0.002698034,-0.0032536325,-0.015248464,0.02832268,-0.0009253666,0.03265303,0.0153594725,-0.009296316,0.0037546174,0.01748552,0.018440893,-0.0013440272,-0.004266834,0.031075153,-0.016378287,-0.008533252,-0.020938974,-0.0006179799,-0.011324175,0.013129271,-0.027681595,-0.012836488,-0.031850614,-0.031083481,-0.0062991446,0.0014365902,-0.031330798,0.018006196,0.036360774,0.012962397,-0.009848979,-0.0074845636,-0.042731013,0.0014276322,0.034473002,0.036104664,-0.10370743,-0.0015518137,-0.013419678,0.009754224,-0.005685581,-0.009154259,-0.022862822,0.010801588,0.014658768,-0.0052282154,0.010147735,0.0053392984,0.009979696,-0.029870598,-0.01988177,0.024591692,-0.012504957,-0.022602597,0.005375365,0.015031157,-0.004837948,-0.016482558,0.0045416094,-0.027559252,-0.0013560915,-0.019620722,0.046645794,-0.01588307,-0.02260588,0.009455474,-0.028812245,-0.014127437,0.008505964,-0.006911671,-0.04734489,0.046493,-0.008898943,-0.00023625286,-0.015100153,-0.004824914,-0.009633642,-0.0032414042,-0.028021999,0.010900696,-0.0011170005,-0.019663418,-0.01874621,0.017370652,0.0024695226,0.008621173,0.03081141,0.026114332,0.017926674,-0.0011154227,0.005542244,0.016953118,-0.005387493,0.011355331,0.026856204,0.0017092278,-0.00014842242,0.0041517103,-0.022350164,0.01113289,-0.02083105,0.009608744,-0.008921372,-0.03956215,0.034212828,-0.011773928,-0.0058232704,-0.002371375,0.016433328,-0.0006402535,-0.016555866,0.0003941968,0.022941241,0.00711231,-0.0025912381,0.010239828,-0.019625517,-0.010583204,-0.026914703,-0.007181246,0.012812667,0.019874755,-0.029847564,0.009168838,-0.012810023,0.038119964,-0.015377875,-0.010474988,-0.036500018,0.001805275,0.003946437,-0.005204519,-0.022780817,0.023207946,-0.01820433,-0.013722324,-0.019829009,0.009578861,-0.0117086675,-0.010169818,-0.016410397,-0.009346306,0.00306948,0.009715872,0.018146617,-0.022666872,-0.0392339,-0.03289118,-0.0082858335,0.011309099,-0.006848682,0.016008047,0.0027986518,0.015863808,0.009048535,0.010730345,0.009266663,0.0015077818,0.004764644,-0.0040004076,0.011742393,-0.009825313,0.023545748,0.00035005913,0.020437688,0.011731509,0.006427183,0.0063775363,-0.001563416,-0.010783054,-0.0046922807,-0.021009872,0.020658059,0.017401805,-0.0109935785,0.020362752,-0.023109203,0.0063139396,-0.008025765,-0.028085224,0.016405296,0.0058846027,-0.0007389502,-0.014918141,-0.0046114577,0.0166454,-0.0065218364,0.008784177,-0.009271867,-0.005436888,0.013074934,0.010569967,0.029127266,0.010922451,-0.006726468,-0.00403121,0.018359493,-0.07160162,0.02337292,0.010985573,-0.015751412,-0.0007356166,0.028491,-0.015725223,0.008132166,-0.008460562,0.02511416,0.010965651,-0.016815785,0.05236204,-0.004273739,-0.0002033407,-0.023173913,-0.026639195,-0.03510443,0.022167793,-0.045536052,-0.00940842,0.0028077664,-0.0062701306,0.0025214888,-0.004325123,-0.015078959,-0.0069782315,0.049981307,0.011354904,-0.001262165,-0.018824005,0.02019266,-0.0011631969,-0.009409887,-0.0039011738,-0.021452028,0.0043238965,-0.01876694,0.01264559,-0.006205424,0.0046225213,0.003282908,0.0066634202,0.01732269,-0.001737406,0.01166269,0.0023896554,0.0065392554,-0.026814807,0.003038784,-0.02337251,0.03497436,0.015370231,-0.012690047,0.0031073801,0.018601114,0.024867153,-0.02007755,-0.018103635,0.00800291,-0.013278037,-0.028083893,-0.0012547179,0.0022508337,-0.015457782,-0.009189279,-0.018395415,-0.0049944194,-0.03697657,-0.0065925033,0.0045601013,0.0046907673,-0.024388801,-0.025147533,-0.0027823148,-0.0027138556,0.0071228235,-0.011519125,-0.0053232727,0.015184811,-0.000758606,-0.0073485,0.011523045,0.023368107,-0.009894667,0.011231932,-0.00017339425,0.0077464813,0.011245552,-0.0151860835,0.016333327,0.028737083,-0.0112017505,0.010135987,-0.018939104,0.009465515,0.02327263,-0.005720081,-0.015535747,-0.010443864,-0.008160456,0.012037206,-0.026562015,-0.0054259836,-0.012890594,-0.018823965,-0.0028795614,-0.03817798,-0.029177696,-0.013618017,0.023090512,-0.023393465,0.015694799,0.010420831,-0.012211287,-0.015762398,0.0041643637,0.006600137,0.0040396117,-0.014735925,-0.019532336,0.017709792,0.0011744517,0.018093009,-0.0022075144,0.0029352766,-0.055406623,-0.013473083,0.010169435,-0.013948142,-0.029599497,0.002886975,-0.008069853,-0.02986992,-0.0098001845,-0.0030255413,0.017021455,0.01956857,0.044559363,0.01743377,0.013786337,0.015304256,0.009476687,-0.016914463,-0.047227874,0.015792185,-0.024156483,-0.003693169,-0.01793592,-0.016163062,-0.050420806,0.011245472,0.006322431,-0.031917457,0.013463138,-0.019011525,-0.010558363,-0.02967075,-0.004618529,-0.015465965,-0.03264735,0.0018378607,-0.005004881,-0.02360339,-0.017749466,0.0006185469,0.007698138,-0.025106726,0.029257456,0.013109419,-0.0011965493,0.006073556,-0.03379027,-0.0035992465,0.008985278,0.004094821,-0.001426213,-0.004745571,-0.013369577,-0.021037431,-0.0012262943,-0.0076607536,0.022501843,-0.04429955,-0.006036996,0.035288308,0.0011101468,0.0041686497,-0.0072882087,-0.0034152996,-0.009088359,0.009548574,-0.0046863155,-0.00065264915,-0.015268776,0.011669528,0.0019025471,-0.027080154,-0.015554134,0.011734901,-0.017403515,-0.026466407,-0.012718846,-0.015587942,0.0066903518,-0.015782915,0.0124585,-0.029692937,0.025008714,-0.00093847397,-0.009243478,0.0035208557,0.019281149,0.029474363,0.008563221,0.028952757,-0.005146965,0.001808441,-0.013925709,-0.007159071,-0.009300419,-0.00374877,-0.00075518637,0.0158846,-0.012692495,0.0077704773,0.011562169,-0.026078785,-0.019677186,-0.023704633,0.026726883,0.013767714,0.015168348,0.00042805518,-0.011704064,-0.012203673,0.012188595,0.020442132,-0.01031819,-0.012139045,0.018462759,0.0016847557,-0.012203195,-0.0007280005,-0.005258365,0.0022793673,0.00652968,0.020953277,-0.012217404,-0.0002507439,1.6877217e-05,0.028206421,-0.004380774,0.0007648178,-0.0133623285,0.0326709,0.015822498,-0.011932528,-0.008784976,0.016110944,0.011031019,-0.009549598,0.02251149,0.01195006,-0.02900716,-0.016751723,-0.0155603755,0.022521388,0.030364018,-0.103442304,-0.0014884023,-0.03976464,0.0029029115,-0.0012199717,-0.037178263,0.0014302866,-0.006180488,-0.026786605,0.011760913,-0.0031628888,-0.003467804,0.019434335,0.00018375524,0.003300196,-0.0003547168,-0.012826197,-0.014009192,0.047704417,-0.016197722,0.018855425,0.017478544,0.02805232,0.027310353,-0.029146297,0.0063781887,0.0027245025,0.029611146,0.017565627,0.022492139,-0.028106315,-0.0083451625,-0.0072958726,0.0053294767,0.011078644,-0.0010684486,-0.005977181,-0.013855023,0.005681738,0.0097495355,-0.015666913,-0.028968573,-0.027688457,-0.006427112,-0.0065873927,-0.011212815,-0.008933269,-0.042332582,-0.0063211923,0.01618056,-0.009244569,-0.03948275,0.00022579962,0.0081319995,0.0061912215,-0.0001715142,-0.030817786,0.015395371,-0.010593981,0.013256706,0.017265607,0.014785061,-0.005713789,0.023657786,-0.003484631,0.026895855,0.026481668,0.027413616,-0.014219017,-0.012077137,-0.018199733,0.011202254,0.0017287781,0.034236684,0.012681691,0.019101946,-0.01050866,0.012261424,0.0072605023,0.014942609,-0.037698653,-0.020080019,-0.08579247,-0.037951823,-0.013825998,-0.0045154477,-0.005086542,-0.007292755,0.01938193,-0.0029271836,0.0063956287,-0.016288646,0.010655824,-0.015662532,0.009800149,-0.040002435,-0.008189205,-0.008994139,0.0026411728,-0.016372457,-0.015854612,-0.021104455,-0.0077714133,0.009501819,0.010715794,0.0016277425,-0.009481872,0.012541058,0.013107738,-0.010342336,4.3659296e-05,-0.0403281,-0.001973033,-0.14257072,0.005636645,0.017371142,-0.010004256,0.012887968,0.02633993,0.022122579,0.0016334874,-0.019674106,-0.03929139,-0.0155174015,-0.011896762,-0.0012635018,-0.002693817,0.009234093,0.10969278,-0.007065529,-0.005212633,0.0018077327,-0.013705678,-0.0030739196,-0.031840224,-0.008475138,0.0036068836,-0.0035238499,0.0038779352,0.016610285,0.020913957,-0.010735668,-0.0049294713,0.006612392,-0.00044787175,-0.019679684,0.015801623,-0.0030155154,0.024590954,0.023308694,-0.01956895,-0.0077069914,0.027545992,-0.0077555235,0.003815778,0.010668842,0.017106164,0.018006496,0.0059660226,-0.013270388,-0.00786663,-0.009843424,-0.032949436,0.0033354114,-0.037508227,0.0011931659,0.0043623406,-0.023236351,-0.009627211,-0.018269345,0.004288806,0.030213946,0.006537906,1.9897794e-05,0.022153655,-0.012403876,0.028361259,0.015491966,0.013088631,0.010450974,0.007442165,-0.0012276646,0.0027680693,-0.0028164436,-0.01814045,-0.022608694,0.01346396,-0.001368374,-0.020730343,-0.026367657,0.011952172,0.03438914,-0.010546976,0.02171645,0.0016420303,-0.018781427,-0.003142302,0.020308562,0.0003826115,-0.00044214464,-0.0032945985,-0.0064925463,-0.0016297392,0.0009582124,0.034959197,-0.0078669675,0.04187284,0.010627694,0.018838907,-0.014338497,0.018615354,0.039071172,-0.0132614,-0.020337228,0.012463996,0.014815789,-0.053757034,0.006638005,-0.003186102,-0.025596973,0.021307154,0.004247839,0.005241244]",{"tags":37,"relatedLang":48,"relatedPosts":52},[38,40,42,44,46],{"name":14,"slug":39},"python",{"name":13,"slug":41},"mistral-ocr",{"name":15,"slug":43},"pdf-ocr",{"name":17,"slug":45},"api-金鑰",{"name":16,"slug":47},"markdown",{"id":27,"slug":49,"title":50,"language":51},"how-to-use-mistral-ocr-with-python-en","How to Use Mistral OCR with Python","en",[53,59,65,71,77,83],{"id":54,"slug":55,"title":56,"cover_image":57,"image_url":57,"created_at":58,"category":26},"dc666c3f-3dc1-43e3-908b-14d7f4cfdac0","ru-he-yong-cuda-oxide-jian-li-rust-gpu-he-xin-zh","如何用 cuda-oxide 建立 Rust GPU 核心","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778976835472-mbgt.png","2026-05-17T00:13:33.173664+00:00",{"id":60,"slug":61,"title":62,"cover_image":63,"image_url":63,"created_at":64,"category":26},"6ca36c73-d147-4134-913d-7e1df080899f","vector-databases-aws-explained-zh","AWS 怎麼看向量資料庫","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778973842531-shap.png","2026-05-16T23:23:31.940718+00:00",{"id":66,"slug":67,"title":68,"cover_image":69,"image_url":69,"created_at":70,"category":26},"4a78a285-c9ab-400d-a4ea-ada7303fd327","how-to-choose-a-vector-database-in-2026-zh","2026 向量資料庫怎麼選","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778972632684-qdj2.png","2026-05-16T23:03:29.482997+00:00",{"id":72,"slug":73,"title":74,"cover_image":75,"image_url":75,"created_at":76,"category":26},"a109dac1-43f3-4a6b-982c-13b59e8f61e9","vibe-research-ai-tools-workflows-zh","Vibe Research：用 AI 加速研究流程","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778904653705-zekc.png","2026-05-16T04:10:33.15767+00:00",{"id":78,"slug":79,"title":80,"cover_image":81,"image_url":81,"created_at":82,"category":26},"cb68bb90-3638-4334-87c7-02580f59877a","aws-repository-wide-security-scanner-matters-zh","為什麼 AWS 的全倉庫安全掃描比更快的 SAST 更重要","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778901047875-n4l9.png","2026-05-16T03:10:24.757504+00:00",{"id":84,"slug":85,"title":86,"cover_image":87,"image_url":87,"created_at":88,"category":26},"7c966206-36f7-4d6b-b2e5-088a4732ede4","why-docker-microvm-sandboxes-ai-agents-zh","為什麼 Docker 的 microVM 沙盒才是 AI 代理的正解","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778893837158-gpxf.png","2026-05-16T01:10:20.668094+00:00",[90,95,100,105,110,115,120,125,130,135],{"id":91,"slug":92,"title":93,"created_at":94},"de769291-4574-4c46-a76d-772bd99e6ec9","googles-biggest-gemini-launches-in-2026-zh","Google 2026 最大 Gemini 盤點","2026-03-26T07:26:39.21072+00:00",{"id":96,"slug":97,"title":98,"created_at":99},"855cd52f-6fab-46cc-a7c1-42195e8a0de4","surepath-real-time-mcp-policy-controls-zh","SurePath 推出即時 MCP 政策控管","2026-03-26T07:57:40.77233+00:00",{"id":101,"slug":102,"title":103,"created_at":104},"9b19ab54-edef-4dbd-9ce4-a51e4bae4ebb","mcp-in-2026-the-ai-tool-layer-teams-use-zh","2026 年 MCP：團隊真的在用的 AI 工具層","2026-03-26T08:01:46.589694+00:00",{"id":106,"slug":107,"title":108,"created_at":109},"af9c46c3-7a28-410b-9f04-32b3de30a68c","prompting-in-2026-what-actually-works-zh","2026 提示工程，真正有用的是什麼","2026-03-26T08:08:12.453028+00:00",{"id":111,"slug":112,"title":113,"created_at":114},"05553086-6ed0-4758-81fd-6cab24b575e0","garry-tan-open-sources-claude-code-toolkit-zh","Garry Tan 開源 Claude Code 工具包","2026-03-26T08:26:20.068737+00:00",{"id":116,"slug":117,"title":118,"created_at":119},"042a73a2-18a2-433d-9e8f-9802b9559aac","github-ai-projects-to-watch-in-2026-zh","2026 必看 20 個 GitHub AI 專案","2026-03-26T08:28:09.619964+00:00",{"id":121,"slug":122,"title":123,"created_at":124},"a5f94120-ac0d-4483-9a8b-63590071ac6a","claude-code-vs-cursor-2026-zh","Claude Code 與 Cursor 深度對比：202…","2026-03-26T13:27:14.279193+00:00",{"id":126,"slug":127,"title":128,"created_at":129},"0975afa1-e0c7-4130-a20d-d890eaed995e","practical-github-guide-learning-ml-2026-zh","2026 機器學習入門 GitHub 實用指南","2026-03-27T01:16:49.712576+00:00",{"id":131,"slug":132,"title":133,"created_at":134},"bfdb467a-290f-4a80-b3a9-6f081afb6dff","aiml-2026-student-ai-ml-lab-repo-review-zh","AIML-2026：像課綱的學生實驗 Repo","2026-03-27T01:21:51.467798+00:00",{"id":136,"slug":137,"title":138,"created_at":139},"80cabc3e-09fc-4ff5-8f07-b8d68f5ae545","ai-trending-github-repos-and-research-feeds-zh","AI Trending：把 AI 資源收成一張表","2026-03-27T01:31:35.262183+00:00"]