[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-turbovec-rust-vector-index-4gb-10m-docs-zh":3,"article-related-turbovec-rust-vector-index-4gb-10m-docs-zh":31,"series-tools-3aa7ba61-2181-4ca8-be11-b26bf62899d1":83},{"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":23,"views":27,"created_at":28,"published_at":29,"topic_cluster_id":30},"3aa7ba61-2181-4ca8-be11-b26bf62899d1","turbovec-rust-vector-index-4gb-10m-docs-zh","TurboVec：Rust 向量索引把 1,000 萬文件壓到 4GB","\u003Cp data-speakable=\"summary\">TurboVec 是一個以 \u003Ca href=\"\u002Ftag\u002Frust\">Rust\u003C\u002Fa> 實作的向量索引，搭配 Python 綁定，可把 1,000 萬文件壓到 4GB 記憶體，並支援過濾搜尋。\u003C\u002Fp>\u003Cp>這個專案來自 \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FRyanCodrai\u002Fturbovec\" target=\"_blank\" rel=\"noopener\">RyanCodrai\u002Fturbovec\u003C\u002Fa>，\u003Ca href=\"\u002Fnews\u002Fwhy-defi-technologies-still-speculation-not-core-holding-zh\">核心\u003C\u002Fa>採用 \u003Ca href=\"\u002Ftag\u002Fgoogle\">Google\u003C\u002Fa> Research 的 \u003Ca href=\"\u002Ftag\u002Fturboquant\">TurboQuant\u003C\u002Fa> \u003Ca href=\"\u002Fnews\u002F5-ways-to-track-dfdv-stock-on-yahoo-finance-zh\">方法\u003C\u002Fa>。作者公布的數字很直接：同一批資料若用 float32，記憶體需求是 31GB；用 TurboVec，則降到 4GB。\u003C\u002Fp>\u003Cp>它不是只做壓縮，還把索引、查詢與持久化一起包進去。對需要在本機或 VPC 裡跑 \u003Ca href=\"\u002Ftag\u002Frag\">RAG\u003C\u002Fa> 的團隊來說，這種設計比把資料丟到託管向量服務更容易控管。\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>項目\u003C\u002Fth>\u003Cth>數值\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Corpus size\u003C\u002Ftd>\u003Ctd>10 million documents\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>RAM with float32\u003C\u002Ftd>\u003Ctd>31 GB\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>RAM with TurboVec\u003C\u002Ftd>\u003Ctd>4 GB\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Repository stars\u003C\u002Ftd>\u003Ctd>3.8k\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Forks\u003C\u002Ftd>\u003Ctd>347\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Commits\u003C\u002Ftd>\u003Ctd>144\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Benchmark speedup on ARM\u003C\u002Ftd>\u003Ctd>12–20% over FAISS FastScan\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Benchmark result on x86\u003C\u002Ftd>\u003Ctd>1–6% faster on 4-bit configs\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>發生了什麼\u003C\u002Fh2>\u003Cp>TurboVec 把 TurboQuant 做成可直接落地的本機索引，支援線上 ingest，不需要先做訓練，也不用等資料量變動後重建整個索引。這對資料持續新增的應用很實用，因為索引維護成本會直接影響上線速度。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780156970637-qrva.png\" alt=\"TurboVec：Rust 向量索引把 1,000 萬文件壓到 4GB\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>專案同時提供兩種主要介面：\u003Ccode>TurboQuantIndex\u003C\u002Fcode> 和 \u003Ccode>IdMapIndex\u003C\u002Fcode>。前者偏向快速建立與查詢，後者則讓外部 ID 保持穩定，方便把向量索引接到既有資料庫或應用層。\u003C\u002Fp>\u003Cp>它的另一個重點是把過濾搜尋放進 SIMD kernel 裡。使用者可以在查詢時傳入 allowlist 或 slot bitmask，系統只從允許集合裡回傳前 \u003Ccode>k\u003C\u002Fcode> 筆結果，減少多撈與多算的浪費。\u003C\u002Fp>\u003Cp>專案頁面也列出一串整合目標，包括 \u003Ca href=\"https:\u002F\u002Fwww.langchain.com\u002F\" target=\"_blank\" rel=\"noopener\">LangChain\u003C\u002Fa>、\u003Ca href=\"https:\u002F\u002Fwww.llamaindex.ai\u002F\" target=\"_blank\" rel=\"noopener\">LlamaIndex\u003C\u002Fa>、\u003Ca href=\"https:\u002F\u002Fhaystack.deepset.ai\u002F\" target=\"_blank\" rel=\"noopener\">Haystack\u003C\u002Fa> 與 \u003Ca href=\"https:\u002F\u002Fdocs.agno.com\u002F\" target=\"_blank\" rel=\"noopener\">Agno\u003C\u002Fa>。這代表它不是只面向底層工程師，也想直接進入現成的 RAG 工具鏈。\u003C\u002Fp>\u003Cul>\u003Cli>Rust core with Python bindings\u003C\u002Fli>\u003Cli>Online ingest, no separate train phase\u003C\u002Fli>\u003Cli>Filtered search with allowlists or bitmasks\u003C\u002Fli>\u003Cli>Local-only use for air-gapped or VPC deployments\u003C\u002Fli>\u003Cli>Adapters for LangChain, LlamaIndex, Haystack, and Agno\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>為什麼重要\u003C\u002Fh2>\u003Cp>對開發者來說，TurboVec 的價值在於把向量搜尋的記憶體門檻拉低。這會直接影響小型 GPU、CPU-only 主機，甚至是邊緣設備能不能跑得動向量檢索。\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780156992740-75a2.png\" alt=\"TurboVec：Rust 向量索引把 1,000 萬文件壓到 4GB\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>它也把「資料不能出域」這件事變得更好處理。若團隊在金融、醫療或內部知識庫場景工作，本機索引加上 Python 綁定，通常比把嵌入向量送去第三方服務更容易過資安與合規審查。\u003C\u002Fp>\u003Cp>\u003Ca href=\"\u002Ftag\u002Fbenchmark\">Benchmark\u003C\u002Fa> 方面，TurboVec 主打在 ARM 上比 \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffaiss\" target=\"_blank\" rel=\"noopener\">FAISS\u003C\u002Fa> 的 FastScan 快 12–20%，在 x86 的 4-bit 配置也有 1–6% 優勢。這些數字還需要更多\u003Ca href=\"\u002Fnews\u002Fbis-tokenization-real-value-payments-zh\">真實\u003C\u002Fa>工作負載驗證，但已經足以讓正在比較 FAISS、託管向量庫和自建索引的團隊把它列入測試清單。\u003C\u002Fp>\u003Cp>真正的問題是：當資料規模、過濾條件和查詢型態變複雜後，TurboVec 的壓縮率與召回率能不能同時守住。若答案是可以，很多 RAG 專案就不必再用記憶體換便利。\u003C\u002Fp>","TurboVec 用 Rust 與 TurboQuant 把 1,000 萬文件的向量索引壓到 4GB，並提供 Python 綁定、過濾搜尋與本機部署能力，主打比 FAISS 更省記憶體。","github.com","https:\u002F\u002Fgithub.com\u002FRyanCodrai\u002Fturbovec",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1780156970637-qrva.png","tools","zh","0d39c74b-f225-4dab-af04-d3fafccb3221",[17,18,19,20,21,22],"TurboVec","Rust","向量索引","RAG","FAISS","TurboQuant",[24,25,26],"10M 文件可壓到 4GB，明顯降低本機向量檢索的記憶體成本。","支援線上 ingest、持久化與過濾搜尋，適合持續成長的 RAG 系統。","對重視資料本地化、合規與成本控制的團隊，這是值得測的 FAISS 替代方案。",3,"2026-05-30T16:02:25.531134+00:00","2026-05-30T16:02:25.525+00:00","c3c88dd2-a940-438a-b359-0e5a24562273",{"tags":32,"relatedLang":42,"relatedPosts":46},[33,35,37,38,40],{"name":18,"slug":34},"rust",{"name":20,"slug":36},"rag",{"name":19,"slug":19},{"name":21,"slug":39},"faiss",{"name":17,"slug":41},"turbovec",{"id":15,"slug":43,"title":44,"language":45},"turbovec-rust-vector-index-4gb-10m-docs-en","TurboVec: Rust vector index cuts 10M docs to 4GB","en",[47,53,59,65,71,77],{"id":48,"slug":49,"title":50,"cover_image":51,"image_url":51,"created_at":52,"category":13},"5656a6ab-9e07-41be-9cea-3440fb8846e2","nvidia-lg-ai-collaboration-playbook-zh","Nvidia 和 LG 把 AI 合作變成模板","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781056994999-8eng.png","2026-06-10T02:02:46.590133+00:00",{"id":54,"slug":55,"title":56,"cover_image":57,"image_url":57,"created_at":58,"category":13},"e48be66d-d7de-419e-b5fd-805f0784ef15","ollama-best-free-ai-path-2026-zh","Ollama 是 2026 年真正適合工作的免費 AI 路徑","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781056077878-11pc.png","2026-06-10T01:47:24.632993+00:00",{"id":60,"slug":61,"title":62,"cover_image":63,"image_url":63,"created_at":64,"category":13},"9b53427c-8c2a-4960-a773-f14d4528caae","awesome-production-ml-turns-chaos-into-stack-zh","這份 MLOps 清單把混亂拆成堆疊","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781055220958-dmar.png","2026-06-10T01:33:14.850634+00:00",{"id":66,"slug":67,"title":68,"cover_image":69,"image_url":69,"created_at":70,"category":13},"d5af1522-28aa-4cfb-8779-1ecf168bc0b5","bentoml-turns-model-serving-into-python-apis-zh","BentoML 把模型服務變成 Python API","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781054310299-c1gm.png","2026-06-10T01:17:56.193093+00:00",{"id":72,"slug":73,"title":74,"cover_image":75,"image_url":75,"created_at":76,"category":13},"63d8b456-ad6b-475e-86e9-d4677ca226aa","magenta-realtime-2-score-inside-daw-zh","Magenta RealTime 2 讓你在 DAW 裡即時改曲","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781046204038-8tox.png","2026-06-09T23:02:55.9651+00:00",{"id":78,"slug":79,"title":80,"cover_image":81,"image_url":81,"created_at":82,"category":13},"f60261ff-a42e-4cfb-9f90-97785e633289","open-source-ai-tools-beat-claude-paid-tiers-zh","開源 AI 工具在價值上已經贏過 Claude 付費方案","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1781045266035-on7t.png","2026-06-09T22:47:20.195939+00:00",[84,89,94,99,104,109,114,119,124,129],{"id":85,"slug":86,"title":87,"created_at":88},"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":90,"slug":91,"title":92,"created_at":93},"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":95,"slug":96,"title":97,"created_at":98},"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":100,"slug":101,"title":102,"created_at":103},"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":105,"slug":106,"title":107,"created_at":108},"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":110,"slug":111,"title":112,"created_at":113},"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":115,"slug":116,"title":117,"created_at":118},"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":120,"slug":121,"title":122,"created_at":123},"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":125,"slug":126,"title":127,"created_at":128},"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",{"id":130,"slug":131,"title":132,"created_at":133},"3ce6e6e2-bac5-463e-9f8d-45caabcc61f7","awesome-ai-for-science-research-tools-map-zh","AI 科研工具清單，開始像地圖了","2026-03-27T01:46:50.521945+00:00"]