[{"data":1,"prerenderedAt":2463},["ShallowReactive",2],{"header-counts":3,"footer-counts":6,"alt-main-lm-studio":8,"alt-list-lm-studio":562},{"tools":4,"reviews":5},98,27,{"tools":4,"reviews":5,"playbooks":7,"news":5},23,{"id":9,"title":10,"alternatives":11,"api_compatible":16,"body":17,"category":492,"chinese_friendly":479,"cover":493,"description":494,"domestic":495,"extension":496,"faq":497,"free":510,"github":16,"languages":511,"lastVerified":514,"meta":515,"models":16,"navigation":510,"notSuitable":16,"opensource":495,"path":516,"pillar":517,"platforms":518,"priceTable":522,"pricing":531,"published":532,"relatedPlaybooks":533,"relatedReviews":16,"score":536,"self_host":510,"seo":539,"seoTitle":540,"slug":541,"sources":542,"stem":550,"suitable":16,"tagline":551,"tags":552,"updated":545,"verdict":559,"website":560,"__hash__":561},"tools\u002Ftools\u002Fcoding\u002Flocal\u002Flm-studio.md","LM Studio",[12,13,14,15],"coding\u002Flocal\u002Follama","coding\u002Flocal\u002Fopen-webui","coding\u002Flocal\u002Fcherry-studio","coding\u002Flocal\u002Flobe-chat",null,{"type":18,"value":19,"toc":477},"minimark",[20,25,34,37,40,100,103,117,123,127,132,149,154,171,174,201,204,358,361,393,397,420,423,450,453],[21,22,24],"h2",{"id":23},"tldr","TL;DR",[26,27,28,29,33],"p",{},"LM Studio 是 Windows \u002F macOS \u002F Linux 桌面应用，让你像浏览 App Store 一样发现、下载、运行本地大模型（GGUF \u002F MLX 格式）。底层基于 llama.cpp + MLX，Mac M 系列原生优化。0.3+ 起新增 Headless 模式 + ",[30,31,32],"code",{},"lms"," CLI，可在服务器跑 OpenAI 兼容 API（默认 :1234）。个人 \u002F 评估完全免费，商用咨询。",[26,35,36],{},"适合：本地 LLM 入门 \u002F 评估、Mac 用户、需要 GUI 调参 \u002F 模型比较、想给 IDE \u002F 应用接本地 OpenAI 兼容 endpoint 的开发者。不适合：多用户并发生产服务（用 vLLM）、嵌入式 \u002F 边缘部署（用 llama.cpp）、纯 CLI 工作流（用 Ollama）。",[21,38,39],{"id":39},"核心能力",[41,42,43,51,57,63,73,82,88,94],"ul",{},[44,45,46,50],"li",{},[47,48,49],"strong",{},"模型浏览器","：内置 Hugging Face 检索，按 GGUF \u002F MLX \u002F 大小筛选、一键下载",[44,52,53,56],{},[47,54,55],{},"聊天界面","：System Prompt \u002F temperature \u002F top-p \u002F context size 可视化调参",[44,58,59,62],{},[47,60,61],{},"多模型并存 \u002F 切换","：同时加载多模型在不同会话中比较",[44,64,65,68,69,72],{},[47,66,67],{},"OpenAI 兼容 Local Server","：",[30,70,71],{},"http:\u002F\u002Flocalhost:1234\u002Fv1","，任何 SDK 即接即用",[44,74,75,68,78,81],{},[47,76,77],{},"Headless \u002F CLI",[30,79,80],{},"lms server start --port 1234","，无 GUI 可跑",[44,83,84,87],{},[47,85,86],{},"PDF \u002F 文档对话","：内置基础 RAG，丢文件就能聊",[44,89,90,93],{},[47,91,92],{},"MLX 原生支持（Mac）","：M1+ 上比 GGUF + Metal 快 30–50%",[44,95,96,99],{},[47,97,98],{},"持续批处理","：Codersera 2026 测得 50–90 tok\u002Fs（消费级 GPU + 中等模型）",[21,101,102],{"id":102},"价格",[41,104,105,111],{},[44,106,107,110],{},[47,108,109],{},"个人 \u002F 评估","：免费，全功能可用",[44,112,113,116],{},[47,114,115],{},"商用","：邮件 \u002F 官网联系 LM Studio 团队",[118,119,120],"blockquote",{},[26,121,122],{},"模型本身免费（开源权重），LM Studio 不抽水任何 token 费用。",[21,124,126],{"id":125},"实测mac-m2-pro-qwen3-coder-7b-gguf-q4_k_m","实测（Mac M2 Pro + Qwen3-Coder-7B GGUF Q4_K_M）",[26,128,129],{},[47,130,131],{},"亮点：",[41,133,134,137,140,143,146],{},[44,135,136],{},"模型浏览器极舒服：搜「qwen3-coder」直接列出 GGUF + MLX 各 quant，标硬件兼容度",[44,138,139],{},"加载 7B Q4 模型 \u003C 3 秒，生成 ~75 tok\u002Fs",[44,141,142],{},"Local Server 开了 Cursor 直接接 baseURL → 本地代码补全零成本",[44,144,145],{},"MLX 版同模型 ~110 tok\u002Fs，差距显著",[44,147,148],{},"多窗口加载 2 个模型并排测，调 prompt 直观",[26,150,151],{},[47,152,153],{},"踩坑：",[41,155,156,159,162,165,168],{},[44,157,158],{},"模型库依赖 Hugging Face，国内访问要镜像 \u002F 代理",[44,160,161],{},"GPU 显存吃满后会自动 offload 到 CPU，无提示就慢下来",[44,163,164],{},"Headless 模式相对 Ollama 偏新，文档稍少",[44,166,167],{},"闭源应用（虽免费），不适合企业合规挂钩",[44,169,170],{},"中文 UI 可用但部分菜单仍英文",[21,172,173],{"id":173},"上手",[175,176,177,180,183,186,189,196],"ol",{},[44,178,179],{},"lmstudio.ai 下载（Mac \u002F Windows \u002F Linux）",[44,181,182],{},"打开 → Discover 标签 → 搜模型（如 qwen3-coder、deepseek-v3 GGUF\u002FMLX）→ Download",[44,184,185],{},"Chat 标签 → 选模型 → 调参聊天",[44,187,188],{},"Local Server 标签 → Start Server → 默认端口 1234",[44,190,191,192,195],{},"在你的应用里：",[30,193,194],{},"baseURL = \"http:\u002F\u002Flocalhost:1234\u002Fv1\"","，API Key 任意",[44,197,198,199],{},"Headless：",[30,200,80],{},[21,202,203],{"id":203},"对比",[205,206,207,228],"table",{},[208,209,210],"thead",{},[211,212,213,217,219,222,225],"tr",{},[214,215,216],"th",{},"维度",[214,218,10],{},[214,220,221],{},"Ollama",[214,223,224],{},"Open WebUI",[214,226,227],{},"llama.cpp",[229,230,231,249,266,282,297,312,327,342],"tbody",{},[211,232,233,237,240,243,246],{},[234,235,236],"td",{},"形态",[234,238,239],{},"GUI + CLI",[234,241,242],{},"CLI Daemon",[234,244,245],{},"Docker UI",[234,247,248],{},"二进制",[211,250,251,254,257,260,263],{},[234,252,253],{},"模型浏览",[234,255,256],{},"✅ 内置",[234,258,259],{},"CLI pull",[234,261,262],{},"无",[234,264,265],{},"手动",[211,267,268,271,274,277,280],{},[234,269,270],{},"参数调优 GUI",[234,272,273],{},"✅",[234,275,276],{},"❌",[234,278,279],{},"部分",[234,281,276],{},[211,283,284,287,290,293,295],{},[234,285,286],{},"OpenAI 兼容 API",[234,288,289],{},"✅ :1234",[234,291,292],{},"✅ :11434",[234,294,273],{},[234,296,273],{},[211,298,299,302,304,307,310],{},[234,300,301],{},"MLX (Mac)",[234,303,273],{},[234,305,306],{},"✅ 0.19+",[234,308,309],{},"–",[234,311,309],{},[211,313,314,317,320,322,324],{},[234,315,316],{},"多用户并发",[234,318,319],{},"弱",[234,321,319],{},[234,323,273],{},[234,325,326],{},"中",[211,328,329,332,335,338,340],{},[234,330,331],{},"开源",[234,333,334],{},"闭源（免费）",[234,336,337],{},"MIT",[234,339,337],{},[234,341,337],{},[211,343,344,347,350,353,355],{},[234,345,346],{},"上手难度",[234,348,349],{},"极低",[234,351,352],{},"低",[234,354,326],{},[234,356,357],{},"高",[21,359,360],{"id":360},"避坑",[41,362,363,369,375,381,387],{},[44,364,365,368],{},[47,366,367],{},"国内下模型走镜像","：HF 直连慢 \u002F 卡，配 HF_ENDPOINT=hf-mirror.com",[44,370,371,374],{},[47,372,373],{},"显存爆 ≠ 报错","：GPU 装不下会无声 offload 到 CPU，关注生成速度，必要时降 quant 或换小模型",[44,376,377,380],{},[47,378,379],{},"MLX 优先（Mac M 系列）","：能下 MLX 版就别下 GGUF，速度差距明显",[44,382,383,386],{},[47,384,385],{},"Local Server 暴露要谨慎","：默认 0.0.0.0 + 无鉴权，对外开放前加反代 + Bearer",[44,388,389,392],{},[47,390,391],{},"闭源合规要核","：企业内部使用前查 license；商用必须联系官方",[21,394,396],{"id":395},"适合-不适合","适合 \u002F 不适合",[41,398,399,402,405,408,411,414,417],{},[44,400,401],{},"✅ 本地 LLM 入门 \u002F 评估",[44,403,404],{},"✅ Mac M 系列用户",[44,406,407],{},"✅ 想给 Cursor \u002F Cline 接本地 OpenAI 兼容 endpoint",[44,409,410],{},"✅ 需要 GUI 调参 \u002F 模型比较",[44,412,413],{},"❌ 多用户并发生产服务",[44,415,416],{},"❌ 嵌入式 \u002F 边缘设备",[44,418,419],{},"❌ 强合规 \u002F 必须开源审计",[21,421,422],{"id":422},"相关阅读",[41,424,425,432,438,444],{},[44,426,427],{},[428,429,431],"a",{"href":430},"\u002Ftools\u002Fcoding\u002Flocal\u002Follama","Ollama 评测",[44,433,434],{},[428,435,437],{"href":436},"\u002Ftools\u002Fcoding\u002Flocal\u002Fopen-webui","Open WebUI 评测",[44,439,440],{},[428,441,443],{"href":442},"\u002Ftools\u002Fcoding\u002Flocal\u002Fcherry-studio","Cherry Studio 评测",[44,445,446],{},[428,447,449],{"href":448},"\u002Fplaybook\u002Fonboarding\u002Fclaude-code-getting-started","Claude Code 上手 Playbook",[21,451,452],{"id":452},"来源",[175,454,455,463,470],{},[44,456,457,458],{},"LM Studio 官网 ",[428,459,460],{"href":460,"rel":461},"https:\u002F\u002Flmstudio.ai\u002F",[462],"nofollow",[44,464,465,466],{},"Codersera — LM Studio Complete Guide 2026 ",[428,467,468],{"href":468,"rel":469},"https:\u002F\u002Fcodersera.com\u002Fblog\u002Flm-studio-complete-guide-2026\u002F",[462],[44,471,472,473],{},"Codersera — Ollama vs LM Studio vs vLLM vs llama.cpp vs MLX 2026 ",[428,474,475],{"href":475,"rel":476},"https:\u002F\u002Fcodersera.com\u002Fblog\u002Follama-vs-lm-studio-vs-vllm-vs-llama-cpp-vs-mlx-2026\u002F",[462],{"title":478,"searchDepth":479,"depth":479,"links":480},"",3,[481,483,484,485,486,487,488,489,490,491],{"id":23,"depth":482,"text":24},2,{"id":39,"depth":482,"text":39},{"id":102,"depth":482,"text":102},{"id":125,"depth":482,"text":126},{"id":173,"depth":482,"text":173},{"id":203,"depth":482,"text":203},{"id":360,"depth":482,"text":360},{"id":395,"depth":482,"text":396},{"id":422,"depth":482,"text":422},{"id":452,"depth":482,"text":452},"local","\u002Fimg\u002Ftools\u002Flm-studio.webp","LM Studio 真实评测：跨平台桌面应用，运行本地 GGUF \u002F MLX 大模型。50–90 tok\u002Fs 持续批处理、OpenAI 兼容本地 API（默认端口 1234）、Headless 模式、Mac \u002F Win 双端。对个人开发者免费，企业咨询。",false,"md",[498,501,504,507],{"q":499,"a":500},"和 Ollama 怎么选？","LM Studio 是 GUI 优先（模型浏览器 + 参数面板 + 聊天界面），适合个人 \u002F 评估 \u002F 上手。Ollama 是 CLI \u002F Daemon 优先（后台跑 + REST API），适合应用嵌入 \u002F 脚本调用。两者都基于 llama.cpp，在 Mac M 系列上都已用 MLX。",{"q":502,"a":503},"支持 MLX 吗？","支持。Mac M1+ 上可加载 MLX 格式模型，速度比 GGUF + Metal 快 30–50%。模型搜索时筛选 MLX 即可。",{"q":505,"a":506},"OpenAI 兼容 API 怎么用？","开 Local Server → 默认端口 1234 → `http:\u002F\u002Flocalhost:1234\u002Fv1`。任何 OpenAI SDK 把 baseURL 改这个就能跑本地模型，零代码改动。",{"q":508,"a":509},"Headless 模式？","0.3+ 起支持 `lms server start` CLI 启动后台服务，无 GUI 即可跑 OpenAI 兼容 API，适合服务器 \u002F SSH 场景。",true,[512,513],"en","zh","2026-08-02",{},"\u002Ftools\u002Fcoding\u002Flocal\u002Flm-studio","coding",[519,520,521],"windows","macos","linux",[523,527],{"plan":109,"price":524,"features":525,"notes":526},"免费","全功能 GUI + Headless API + GGUF\u002FMLX","供个人 \u002F 评估使用",{"plan":115,"price":528,"features":529,"notes":530},"联系咨询","团队部署 \u002F 商用 license","邮件 \u002F 官网联系","免费（个人 \u002F 评估） \u002F 企业 \u002F 商用咨询","2026-06-19",[534,535],"onboarding\u002Frag-pipeline-build","onboarding\u002Fclaude-code-getting-started",{"power":537,"ux":538,"price":538,"cn_support":479,"stability":537},4,5,{"title":10,"description":494},"LM Studio 评测 2026：本地运行开源大模型，图形化界面，AI 模型管理","coding\u002Flocal\u002Flm-studio",[543,546,548],{"name":544,"url":460,"accessed":545},"LM Studio 官网","2026-06-24",{"name":547,"url":468,"accessed":545},"Codersera — LM Studio Complete Guide 2026",{"name":549,"url":475,"accessed":545},"Codersera — Ollama vs LM Studio vs vLLM 2026","tools\u002Fcoding\u002Flocal\u002Flm-studio","本地 LLM 的 GUI 首选——模型浏览器 + GGUF\u002FMLX 推理 + OpenAI 兼容 API + Mac 原生优化",[492,553,554,555,556,557,558],"gui","gguf","mlx","llama-cpp","mac","openai-compatible","Mac \u002F Windows 桌面本地 LLM 的 GUI 首选——上手最快、模型浏览最舒服、自带 OpenAI 兼容 API。批量服务 \u002F 多用户场景用 vLLM；纯 CLI \u002F 嵌入应用走 Ollama。","https:\u002F\u002Flmstudio.ai","Pj3jNb1Z4S55e91UkW1ZMgI86ps_Wm7yCc7zeXPF05U",[563,1050,1537,1980],{"id":564,"title":221,"alternatives":565,"api_compatible":566,"body":581,"category":492,"chinese_friendly":479,"cover":1003,"description":1004,"domestic":495,"extension":496,"faq":1005,"free":510,"github":1018,"languages":1019,"lastVerified":514,"meta":1020,"models":16,"navigation":510,"notSuitable":16,"opensource":510,"path":430,"pillar":517,"platforms":1021,"priceTable":1023,"pricing":1028,"published":532,"relatedPlaybooks":1029,"relatedReviews":16,"score":1030,"self_host":510,"seo":1031,"seoTitle":1032,"slug":12,"sources":1033,"stem":1039,"suitable":16,"tagline":1040,"tags":1041,"updated":545,"verdict":1047,"website":1048,"__hash__":1049},"tools\u002Ftools\u002Fcoding\u002Flocal\u002Follama.md",[541,13,14,15],[567,568,569,570,571,572,573,574,575,576,577,578,579,221,580],"OpenAI","Anthropic","Google","Grok","Mistral","Cohere","阿里通义","百度文心","腾讯混元","Moonshot Kimi","字节豆包","DeepSeek","智谱 GLM","Hugging Face",{"type":18,"value":582,"toc":991},[583,585,592,595,597,665,667,670,674,678,698,702,733,735,769,771,886,888,920,922,945,947,968,970],[21,584,24],{"id":23},[26,586,587,588,591],{},"Ollama 是本地 LLM 的 Daemon 事实标准——后台跑、暴露 REST API（11434）+ CLI、Modelfile 配置、GGUF 一站式。MIT 开源，跨 Win \u002F Mac \u002F Linux。0.19+ 起 Mac M 系列底层切 MLX 推理。模型库覆盖 Llama \u002F Qwen \u002F DeepSeek \u002F Gemma \u002F Mistral 等主流开源模型，",[30,589,590],{},"ollama pull"," 一键拉。",[26,593,594],{},"适合：给 Cursor \u002F Cline \u002F Continue \u002F Open WebUI 接本地 OpenAI 兼容 endpoint、个人 \u002F 评估 \u002F 原型、嵌入应用、自动化脚本。不适合：GUI 偏好用户（用 LM Studio）、多用户并发生产服务（用 vLLM）、模型浏览 \u002F 调参界面（用 LM Studio）。",[21,596,39],{"id":39},[41,598,599,605,613,619,626,641,647,653,659],{},[44,600,601,604],{},[47,602,603],{},"后台 Daemon","：开机自启，应用调用零延迟",[44,606,607,68,610],{},[47,608,609],{},"CLI",[30,611,612],{},"ollama pull \u002F run \u002F list \u002F show \u002F create \u002F serve",[44,614,615,618],{},[47,616,617],{},"Modelfile","：类 Dockerfile 注册任意 GGUF，配 SYSTEM \u002F PARAMETER \u002F TEMPLATE",[44,620,621,68,623],{},[47,622,286],{},[30,624,625],{},"http:\u002F\u002Flocalhost:11434\u002Fv1\u002Fchat\u002Fcompletions",[44,627,628,68,631,634,635,634,638],{},[47,629,630],{},"原生 API",[30,632,633],{},"\u002Fapi\u002Fchat","、",[30,636,637],{},"\u002Fapi\u002Fgenerate",[30,639,640],{},"\u002Fapi\u002Fembeddings",[44,642,643,646],{},[47,644,645],{},"模型库","：官方注册表内置 Llama \u002F Qwen \u002F DeepSeek \u002F Gemma \u002F Mistral \u002F GPT-OSS 等",[44,648,649,652],{},[47,650,651],{},"MLX 加速（Mac）","：0.19+ 起 M 系列自动用 MLX",[44,654,655,658],{},[47,656,657],{},"量化","：默认 Q4_K_M、支持 Q5 \u002F Q8 \u002F FP16",[44,660,661,664],{},[47,662,663],{},"跨平台","：Win \u002F Mac \u002F Linux 安装包，Docker 官方镜像",[21,666,102],{"id":102},[26,668,669],{},"完全免费、MIT 开源、商用免费。",[21,671,673],{"id":672},"实测m2-pro-qwen3-coder-7b-q4","实测（M2 Pro + Qwen3-Coder-7B Q4）",[26,675,676],{},[47,677,131],{},[41,679,680,686,689,692,695],{},[44,681,682,685],{},[30,683,684],{},"ollama run qwen3-coder:7b"," 一行起飞，3 秒进交互",[44,687,688],{},"REST API 配 Cursor \u002F Cline \u002F Continue 几乎全工具开箱即用",[44,690,691],{},"Modelfile 写自定义编码助手（low temperature + system prompt + 16K context）几分钟搞定",[44,693,694],{},"多模型并存，按需切换，内存占用合理",[44,696,697],{},"Mac M 系列 MLX 后比旧 GGUF 模式快显著",[26,699,700],{},[47,701,153],{},[41,703,704,714,720,727,730],{},[44,705,706,707,710,711],{},"默认 ",[30,708,709],{},"num_ctx"," 偏小（2048），跑长上下文要在 Modelfile 加 ",[30,712,713],{},"PARAMETER num_ctx 16384",[44,715,716,717],{},"模型默认走 0.0.0.0:11434 ↔ Docker 容器互访要 ",[30,718,719],{},"--add-host=host.docker.internal:host-gateway",[44,721,722,723,726],{},"国内 ",[30,724,725],{},"ollama.com\u002Flibrary"," 下载偶有慢，可手动 HF 下 GGUF + Modelfile 自建",[44,728,729],{},"多用户并发吞吐显著低于 vLLM",[44,731,732],{},"没有 GUI，模型浏览 \u002F 参数面板要走 LM Studio \u002F Open WebUI 配合",[21,734,173],{"id":173},[175,736,737,743,749,754,760,766],{},[44,738,739,742],{},[30,740,741],{},"curl -fsSL https:\u002F\u002Follama.ai\u002Finstall.sh | sh","（Mac \u002F Linux）；Windows winget",[44,744,745,748],{},[30,746,747],{},"ollama pull qwen3-coder:7b","（按需换模型）",[44,750,751,753],{},[30,752,684],{}," 直接聊",[44,755,756,757],{},"应用接入：baseURL = ",[30,758,759],{},"http:\u002F\u002Flocalhost:11434\u002Fv1",[44,761,762,763],{},"自定义：写 Modelfile → ",[30,764,765],{},"ollama create my-coder -f Modelfile",[44,767,768],{},"进阶：装 Open WebUI 做前端 \u002F 多人共享",[21,770,203],{"id":203},[205,772,773,788],{},[208,774,775],{},[211,776,777,779,781,783,786],{},[214,778,216],{},[214,780,221],{},[214,782,10],{},[214,784,785],{},"vLLM",[214,787,227],{},[229,789,790,806,818,831,844,860,872],{},[211,791,792,794,797,800,803],{},[234,793,236],{},[234,795,796],{},"CLI + Daemon",[234,798,799],{},"GUI + Headless",[234,801,802],{},"Python Server",[234,804,805],{},"C++ 二进制",[211,807,808,810,812,814,816],{},[234,809,173],{},[234,811,349],{},[234,813,349],{},[234,815,326],{},[234,817,357],{},[211,819,820,822,824,827,829],{},[234,821,253],{},[234,823,609],{},[234,825,826],{},"✅ GUI",[234,828,262],{},[234,830,262],{},[211,832,833,836,838,840,842],{},[234,834,835],{},"OpenAI 兼容",[234,837,292],{},[234,839,289],{},[234,841,273],{},[234,843,273],{},[211,845,846,849,852,855,858],{},[234,847,848],{},"多用户吞吐",[234,850,851],{},"弱（~40 tok\u002Fs）",[234,853,854],{},"中（50–90）",[234,856,857],{},"强（800–12500）",[234,859,326],{},[211,861,862,864,866,868,870],{},[234,863,301],{},[234,865,306],{},[234,867,273],{},[234,869,279],{},[234,871,309],{},[211,873,874,876,878,881,884],{},[234,875,331],{},[234,877,337],{},[234,879,880],{},"闭源",[234,882,883],{},"Apache 2.0",[234,885,337],{},[21,887,360],{"id":360},[41,889,890,896,902,908,914],{},[44,891,892,895],{},[47,893,894],{},"num_ctx 一定要设","：默认 2K 太小，跑代码 \u002F 长文档要 16K+",[44,897,898,901],{},[47,899,900],{},"Modelfile 模板别漏 TEMPLATE","：错的 chat template 会让模型输出乱码 \u002F 不停",[44,903,904,907],{},[47,905,906],{},"KV cache 爆表 = 速度悬崖","：32B 模型 32K 上下文，KV cache 可能 12+ GB，超显存自动 offload 慢 10×",[44,909,910,913],{},[47,911,912],{},"不要 0.0.0.0 直接对公网","：默认无鉴权，对外暴露走反代 + Bearer \u002F mTLS",[44,915,916,919],{},[47,917,918],{},"Mac 让它自动用 MLX","：升 0.19+；不要手动强制 GGUF + Metal",[21,921,396],{"id":395},[41,923,924,927,930,933,936,939,942],{},[44,925,926],{},"✅ 应用 \u002F IDE 接本地模型（Cursor \u002F Cline \u002F Continue）",[44,928,929],{},"✅ 个人 \u002F 评估 \u002F 脚本自动化",[44,931,932],{},"✅ Modelfile 自定义系统 prompt + 参数",[44,934,935],{},"✅ Mac M 系列 MLX 用户",[44,937,938],{},"❌ 多用户并发生产服务（用 vLLM）",[44,940,941],{},"❌ GUI 调参 \u002F 模型浏览（配 LM Studio \u002F Open WebUI）",[44,943,944],{},"❌ 极致单卡吞吐研究（直接 llama.cpp \u002F vLLM）",[21,946,422],{"id":422},[41,948,949,954,958,962],{},[44,950,951],{},[428,952,953],{"href":516},"LM Studio 评测",[44,955,956],{},[428,957,437],{"href":436},[44,959,960],{},[428,961,443],{"href":442},[44,963,964],{},[428,965,967],{"href":966},"\u002Fplaybook\u002Fonboarding\u002Frag-pipeline-build","RAG Pipeline 搭建 Playbook",[21,969,452],{"id":452},[175,971,972,979,986],{},[44,973,974,975],{},"Markaicode — Import GGUF Models into Ollama 2026（2026-05-15）",[428,976,977],{"href":977,"rel":978},"https:\u002F\u002Fmarkaicode.com\u002Fimport-gguf-models-ollama-guide",[462],[44,980,981,982],{},"ComputingForGeeks — Ollama Models Cheat Sheet 2026 ",[428,983,984],{"href":984,"rel":985},"https:\u002F\u002Fcomputingforgeeks.com\u002Follama-models-cheat-sheet",[462],[44,987,472,988],{},[428,989,475],{"href":475,"rel":990},[462],{"title":478,"searchDepth":479,"depth":479,"links":992},[993,994,995,996,997,998,999,1000,1001,1002],{"id":23,"depth":482,"text":24},{"id":39,"depth":482,"text":39},{"id":102,"depth":482,"text":102},{"id":672,"depth":482,"text":673},{"id":173,"depth":482,"text":173},{"id":203,"depth":482,"text":203},{"id":360,"depth":482,"text":360},{"id":395,"depth":482,"text":396},{"id":422,"depth":482,"text":422},{"id":452,"depth":482,"text":452},"\u002Fimg\u002Ftools\u002Follama.webp","Ollama 真实评测：本地 LLM 的事实标准 Daemon，CLI + REST API，模型库 + Modelfile + GGUF 一站式。0.19+ 在 Mac M 系列用 MLX 加速；OpenAI 兼容端点 11434；MIT 开源 + 跨平台。",[1006,1009,1012,1015],{"q":1007,"a":1008},"和 LM Studio 怎么选？","Ollama = Daemon + CLI，开机自启在 11434 端口跑，应用 \u002F IDE 调它最方便。LM Studio = GUI，模型浏览 \u002F 调参 \u002F 聊天体验更好。两者底层都基于 llama.cpp，Mac M 系列上都已切 MLX。",{"q":1010,"a":1011},"Modelfile 是什么？","类 Dockerfile 的模型配置：`FROM .\u002Fxxx.gguf` + PARAMETER \u002F TEMPLATE \u002F SYSTEM。把任意 GGUF 注册成本地模型供调用。`ollama create my-model -f Modelfile`。",{"q":1013,"a":1014},"OpenAI 兼容端点？","`http:\u002F\u002Flocalhost:11434\u002Fv1`。任何 OpenAI SDK 改 baseURL 即用。也可走原生 `\u002Fapi\u002Fchat`、`\u002Fapi\u002Fgenerate`。",{"q":1016,"a":1017},"并发能力？","单用户原型场景顺滑（~40 tok\u002Fs peak），多用户并发明显不如 vLLM（vLLM 的 PagedAttention + 连续批处理高 16–20×）。生产并发选 vLLM。","https:\u002F\u002Fgithub.com\u002Follama\u002Follama",[512],{},[519,520,521,1022],"docker",[1024],{"plan":1025,"price":524,"features":1026,"notes":1027},"开源版","完整 CLI + REST API + Modelfile + 模型库 + MIT 协议","全平台、商用免费","完全免费 + 开源（MIT）",[534,535],{"power":537,"ux":537,"price":538,"cn_support":479,"stability":538},{"title":221,"description":1004},"Ollama 评测 2026：本地运行大模型，开源 AI 模型管理工具，私有化部署指南",[1034,1036,1038],{"name":1035,"url":977,"accessed":545},"Markaicode — Import GGUF 2026",{"name":1037,"url":984,"accessed":545},"ComputingForGeeks — Ollama Cheat Sheet 2026",{"name":549,"url":475,"accessed":545},"tools\u002Fcoding\u002Flocal\u002Follama","本地 LLM 的 Daemon——CLI + REST API 后台跑，给 Cursor \u002F Cline \u002F Open WebUI 接本地模型最低门槛",[492,1042,1043,1044,1045,554,555,558,1046],"daemon","cli","rest-api","modelfile","open-source","本地 LLM 的 Daemon 事实标准，CLI \u002F Modelfile \u002F REST API 三件套配合最广泛。GUI 偏好用户走 LM Studio；多用户并发生产用 vLLM；其他场景几乎默认 Ollama。","https:\u002F\u002Follama.com","yL3ZqN3rlWsImgvBFSYlFraTqj9ki9gSJlMbXVmruyg",{"id":1051,"title":224,"alternatives":1052,"api_compatible":1053,"body":1054,"category":492,"chinese_friendly":537,"cover":1486,"description":1487,"domestic":495,"extension":496,"faq":1488,"free":510,"github":1501,"languages":1502,"lastVerified":514,"meta":1503,"models":16,"navigation":510,"notSuitable":16,"opensource":510,"path":436,"pillar":517,"platforms":1504,"priceTable":1506,"pricing":1515,"published":532,"relatedPlaybooks":1516,"relatedReviews":16,"score":1517,"self_host":510,"seo":1518,"seoTitle":1519,"slug":13,"sources":1520,"stem":1527,"suitable":16,"tagline":1528,"tags":1529,"updated":545,"verdict":1534,"website":1535,"__hash__":1536},"tools\u002Ftools\u002Fcoding\u002Flocal\u002Fopen-webui.md",[15,14,12,541],[567,568,569,570,571,572,573,574,575,576,577,578,579,221,580],{"type":18,"value":1055,"toc":1474},[1056,1058,1061,1064,1066,1128,1130,1133,1137,1141,1169,1173,1197,1199,1233,1235,1357,1359,1402,1404,1427,1429,1449,1451],[21,1057,24],{"id":23},[26,1059,1060],{},"Open WebUI（原 Ollama WebUI）是 MIT 开源、自托管 AI 平台，最常见用法是 Docker 跑起来给 Ollama 套一个 ChatGPT 风格前端。GitHub 126k+ stars、282M+ Docker pulls，事实上的本地 AI 前端首选。支持任意 OpenAI 兼容后端 + RAG 知识库 + 多用户账号 + 工具调用 + MCP-OpenAPI 代理 + 联网搜索 + 语音 + 图像生成。",[26,1062,1063],{},"适合：团队 \u002F 家庭 \u002F 公司部署一份共享、要 Web 端访问、多用户分账号、SearXNG 联网搜索、Confluence \u002F S3 \u002F GitHub 数据源同步。不适合：单人桌面体验（用 Cherry Studio）、零运维 \u002F 不愿碰 Docker。",[21,1065,39],{"id":39},[41,1067,1068,1074,1080,1086,1092,1098,1104,1110,1116,1122],{},[44,1069,1070,1073],{},[47,1071,1072],{},"多模型后端","：Ollama \u002F OpenAI \u002F vLLM \u002F Anthropic \u002F Groq \u002F LocalAI \u002F 任意 OpenAI 兼容",[44,1075,1076,1079],{},[47,1077,1078],{},"多用户 + RBAC","：注册 \u002F 邀请 \u002F 角色权限 \u002F 工作区隔离",[44,1081,1082,1085],{},[47,1083,1084],{},"RAG 知识库","：上传文档 \u002F 网址 \u002F SearXNG 联网搜索 → 向量化 → 对话引用",[44,1087,1088,1091],{},[47,1089,1090],{},"Tools \u002F Functions","：Python 写函数即扩展（联网 \u002F 计算器 \u002F 自定义 API）",[44,1093,1094,1097],{},[47,1095,1096],{},"mcpo","：MCP-to-OpenAPI 代理，任意 MCP 服务器接进来",[44,1099,1100,1103],{},[47,1101,1102],{},"oikb","：知识库同步本地文件夹 \u002F GitHub \u002F S3 \u002F Confluence 等 40+ 源",[44,1105,1106,1109],{},[47,1107,1108],{},"open-terminal \u002F cptr","：给 AI 真实终端 + 文件 + 沙箱执行",[44,1111,1112,1115],{},[47,1113,1114],{},"图像生成","：Stable Diffusion \u002F DALL·E \u002F 自托管接入",[44,1117,1118,1121],{},[47,1119,1120],{},"语音输入 \u002F TTS","：内置",[44,1123,1124,1127],{},[47,1125,1126],{},"企业 LTS","：custom branding + SLA + 长期支持版本（联系销售）",[21,1129,102],{"id":102},[26,1131,1132],{},"完全免费、MIT 开源、商用免费。Enterprise 提供品牌定制 + SLA + LTS。",[21,1134,1136],{"id":1135},"实测ubuntu-2404-ollama-后端-5-人小团队","实测（Ubuntu 24.04 + Ollama 后端 + 5 人小团队）",[26,1138,1139],{},[47,1140,131],{},[41,1142,1143,1150,1153,1160,1163,1166],{},[44,1144,1145,1146,1149],{},"单条 ",[30,1147,1148],{},"docker run"," 五分钟上线",[44,1151,1152],{},"自带的多用户 + 角色权限省去重新搭 Auth",[44,1154,1155,1156,1159],{},"RAG 直传 30 个 PDF 后向量化顺利，对话中 ",[30,1157,1158],{},"#知识库"," 引用准确",[44,1161,1162],{},"mcpo 把 GitHub MCP 服务器接进来，团队对话里直接 issue \u002F PR 操作",[44,1164,1165],{},"模型切换流畅，OpenAI + Ollama 并存",[44,1167,1168],{},"SearXNG 联网搜索给模型实时信息，过时知识截止问题缓解",[26,1170,1171],{},[47,1172,153],{},[41,1174,1175,1178,1184,1191,1194],{},[44,1176,1177],{},"Docker 镜像 ~1.5GB，首次拉取偏慢",[44,1179,706,1180,1183],{},[30,1181,1182],{},"0.0.0.0"," 公网暴露要加 HTTPS + 反代",[44,1185,1186,1187,1190],{},"嵌入模型 ",[30,1188,1189],{},"sentence-transformers"," 中文效果一般，建议换 bge-m3",[44,1192,1193],{},"多用户共享 Ollama 时并发吞吐瓶颈在 Ollama，不在 Open WebUI（生产用 vLLM 后端）",[44,1195,1196],{},"版本升级要看 changelog，部分 minor 含 breaking 改动",[21,1198,173],{"id":173},[175,1200,1201,1207,1214,1217,1220,1223,1226],{},[44,1202,1203,1204],{},"装 Docker → ",[30,1205,1206],{},"docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:\u002Fapp\u002Fbackend\u002Fdata --name open-webui --restart always ghcr.io\u002Fopen-webui\u002Fopen-webui:main",[44,1208,1209,1210,1213],{},"浏览器开 ",[30,1211,1212],{},"http:\u002F\u002Flocalhost:3000"," → 注册第一个账号（管理员）",[44,1215,1216],{},"设置 → Connections → 连接 Ollama \u002F 加 OpenAI Key",[44,1218,1219],{},"Models → Pull \u002F Discover 模型",[44,1221,1222],{},"Workspaces → 建知识库 → 上传文档",[44,1224,1225],{},"Tools → 启用 \u002F 写自定义函数",[44,1227,1228,1229,1232],{},"生产部署：Nginx 反代 + Let's Encrypt + 备份 ",[30,1230,1231],{},"\u002Fapp\u002Fbackend\u002Fdata"," volume",[21,1234,203],{"id":203},[205,1236,1237,1253],{},[208,1238,1239],{},[211,1240,1241,1243,1245,1248,1251],{},[214,1242,216],{},[214,1244,224],{},[214,1246,1247],{},"LobeChat",[214,1249,1250],{},"Cherry Studio",[214,1252,10],{},[229,1254,1255,1270,1284,1298,1312,1327,1343],{},[211,1256,1257,1259,1262,1265,1268],{},[234,1258,236],{},[234,1260,1261],{},"Docker \u002F 桌面",[234,1263,1264],{},"Web + 桌面",[234,1266,1267],{},"桌面",[234,1269,1267],{},[211,1271,1272,1275,1278,1280,1282],{},[234,1273,1274],{},"多用户",[234,1276,1277],{},"✅ 一等",[234,1279,273],{},[234,1281,262],{},[234,1283,262],{},[211,1285,1286,1289,1292,1294,1296],{},[234,1287,1288],{},"RAG",[234,1290,1291],{},"✅ 强 + oikb",[234,1293,273],{},[234,1295,273],{},[234,1297,319],{},[211,1299,1300,1303,1306,1308,1310],{},[234,1301,1302],{},"工具 \u002F MCP",[234,1304,1305],{},"✅ mcpo",[234,1307,273],{},[234,1309,273],{},[234,1311,319],{},[211,1313,1314,1317,1320,1323,1325],{},[234,1315,1316],{},"自托管",[234,1318,1319],{},"✅ Docker \u002F K8s",[234,1321,1322],{},"✅ Docker",[234,1324,262],{},[234,1326,262],{},[211,1328,1329,1332,1335,1338,1341],{},[234,1330,1331],{},"GitHub Stars",[234,1333,1334],{},"126k+",[234,1336,1337],{},"72k+",[234,1339,1340],{},"60k+",[234,1342,309],{},[211,1344,1345,1348,1350,1352,1355],{},[234,1346,1347],{},"开源协议",[234,1349,337],{},[234,1351,337],{},[234,1353,1354],{},"AGPL-3.0",[234,1356,334],{},[21,1358,360],{"id":360},[41,1360,1361,1367,1375,1384,1390,1396],{},[44,1362,1363,1366],{},[47,1364,1365],{},"不要裸 0.0.0.0 + HTTP 暴露公网","：默认无 HTTPS，必上反代 + 强密码 + 速率限制",[44,1368,1369,1374],{},[47,1370,1371,1372,1232],{},"备份 ",[30,1373,1231],{},"：知识库 \u002F 用户 \u002F 对话全在里面",[44,1376,1377,1380,1381,1383],{},[47,1378,1379],{},"中文 RAG 换嵌入模型","：默认 ",[30,1382,1189],{}," 中文一般，配 bge-m3 或硅基流动嵌入 API",[44,1385,1386,1389],{},[47,1387,1388],{},"mcpo 工具范围谨慎","：MCP 给 AI 真实能力，第三方服务器审一遍",[44,1391,1392,1395],{},[47,1393,1394],{},"后端吞吐看 Ollama","：5+ 并发上 vLLM 后端，Ollama 单 worker 会排队",[44,1397,1398,1401],{},[47,1399,1400],{},"升级前看 changelog","：weekly 更新，偶有 breaking",[21,1403,396],{"id":395},[41,1405,1406,1409,1412,1415,1418,1421,1424],{},[44,1407,1408],{},"✅ 团队 \u002F 家庭 \u002F 公司多人共享 AI 平台",[44,1410,1411],{},"✅ 要 Web 端访问 \u002F 移动端兼容",[44,1413,1414],{},"✅ 自托管 \u002F 完全控制数据",[44,1416,1417],{},"✅ MCP \u002F 工具调用刚需",[44,1419,1420],{},"❌ 单人桌面体验（用 Cherry Studio）",[44,1422,1423],{},"❌ 零运维 \u002F 不愿碰 Docker",[44,1425,1426],{},"❌ iOS 原生 App 主力",[21,1428,422],{"id":422},[41,1430,1431,1437,1441,1445],{},[44,1432,1433],{},[428,1434,1436],{"href":1435},"\u002Ftools\u002Fcoding\u002Flocal\u002Flobe-chat","LobeChat 评测",[44,1438,1439],{},[428,1440,443],{"href":442},[44,1442,1443],{},[428,1444,431],{"href":430},[44,1446,1447],{},[428,1448,967],{"href":966},[21,1450,452],{"id":452},[175,1452,1453,1460,1467],{},[44,1454,1455,1456],{},"Open WebUI 官方文档 ",[428,1457,1458],{"href":1458,"rel":1459},"https:\u002F\u002Fdocs.openwebui.com\u002F",[462],[44,1461,1462,1463],{},"Local AI Master — Open WebUI Setup Guide 2026 ",[428,1464,1465],{"href":1465,"rel":1466},"https:\u002F\u002Flocalaimaster.com\u002Fblog\u002Fopen-webui-setup-guide",[462],[44,1468,1469,1470],{},"AIToolDiscovery — Set Up Open-WebUI with Ollama 2026 ",[428,1471,1472],{"href":1472,"rel":1473},"https:\u002F\u002Fwww.aitooldiscovery.com\u002Fhow-to\u002Fsetup-open-webui-ollama",[462],{"title":478,"searchDepth":479,"depth":479,"links":1475},[1476,1477,1478,1479,1480,1481,1482,1483,1484,1485],{"id":23,"depth":482,"text":24},{"id":39,"depth":482,"text":39},{"id":102,"depth":482,"text":102},{"id":1135,"depth":482,"text":1136},{"id":173,"depth":482,"text":173},{"id":203,"depth":482,"text":203},{"id":360,"depth":482,"text":360},{"id":395,"depth":482,"text":396},{"id":422,"depth":482,"text":422},{"id":452,"depth":482,"text":452},"\u002Fimg\u002Ftools\u002Fopen-webui.webp","Open WebUI 2026 真实评测：MIT 开源、自托管 ChatGPT 替代和 Ollama Web 前端。支持 Docker 一行部署、Ollama\u002FOpenAI\u002FvLLM 多后端、RAG 知识库、多用户、联网搜索、工具调用和 MCP-to-OpenAPI，适合团队私有 AI 平台。",[1489,1492,1495,1498],{"q":1490,"a":1491},"Docker 一行命令真的够用吗？","够。`docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:\u002Fapp\u002Fbackend\u002Fdata --name open-webui --restart always ghcr.io\u002Fopen-webui\u002Fopen-webui:main`，5 分钟可上线、能多人注册、能接 Ollama \u002F OpenAI。生产再加反代 + HTTPS + 备份。",{"q":1493,"a":1494},"支持哪些模型后端？","Ollama（首选）+ 任何 OpenAI 兼容 endpoint：OpenAI 官方 \u002F Anthropic（OpenAI 兼容代理）\u002F vLLM \u002F Groq \u002F LocalAI \u002F 自建 baseURL。可同时配多个，对话中切换。",{"q":1496,"a":1497},"RAG \u002F 知识库怎么做？","内置：上传 PDF \u002F DOCX \u002F TXT、网址抓取、SearXNG 联网搜索 → 自动向量化 → 在对话中 `#` 引用知识库。配套 oikb 项目可同步本地文件夹 \u002F GitHub \u002F S3 \u002F Confluence 等 40+ 数据源。",{"q":1499,"a":1500},"MCP 怎么接？","通过 mcpo（官方的 MCP-to-OpenAPI 代理）把任意 MCP 服务器暴露成 OpenAPI 工具，再在 Open WebUI 注册即可。无需写 glue code。","https:\u002F\u002Fgithub.com\u002Fopen-webui\u002Fopen-webui",[512,513],{},[1022,521,520,519,1505],"kubernetes",[1507,1510],{"plan":1025,"price":524,"features":1508,"notes":1509},"全功能 \u002F 多用户 \u002F RAG \u002F Tools \u002F 联网搜索 \u002F MCP-OpenAPI 代理 \u002F Docker \u002F K8s","MIT 协议",{"plan":1511,"price":1512,"features":1513,"notes":1514},"Enterprise","咨询","Custom branding \u002F SLA \u002F LTS 长期支持版本","邮件官方","完全免费（MIT 开源） \u002F Enterprise SLA 联系",[534,535],{"power":538,"ux":537,"price":538,"cn_support":537,"stability":538},{"title":224,"description":1487},"Open WebUI 评测 2026：自托管 ChatGPT 替代，Ollama 前端部署指南",[1521,1523,1525],{"name":1522,"url":1458,"accessed":545},"Open WebUI 官方文档",{"name":1524,"url":1465,"accessed":545},"Local AI Master — Open WebUI Setup Guide 2026",{"name":1526,"url":1472,"accessed":545},"AIToolDiscovery — Open-WebUI with Ollama 2026","tools\u002Fcoding\u002Flocal\u002Fopen-webui","自托管的 ChatGPT 替代：Ollama \u002F OpenAI 兼容、多用户、RAG、126k+ GitHub stars",[492,1530,1022,1531,1532,1533,1046],"self-host","rag","multi-user","ollama","自托管多用户 AI 前端的事实标准。团队 \u002F 家庭 \u002F 公司部署一份共享，多模型聚合 + RAG + 工具调用全有。单机 \u002F 桌面体验首选 Cherry Studio \u002F LobeChat。","https:\u002F\u002Fdocs.openwebui.com","5Y-zy_XMSVV_gsS0aa6V7zk268r2TvaIcssIp-HXM_U",{"id":1538,"title":1250,"alternatives":1539,"api_compatible":16,"body":1540,"category":492,"chinese_friendly":538,"cover":1929,"description":1930,"domestic":510,"extension":496,"faq":1931,"free":510,"github":1944,"languages":1945,"lastVerified":514,"meta":1946,"models":16,"navigation":510,"notSuitable":16,"opensource":510,"path":442,"pillar":517,"platforms":1947,"priceTable":1949,"pricing":1957,"published":532,"relatedPlaybooks":1958,"relatedReviews":16,"score":1960,"self_host":510,"seo":1961,"seoTitle":1962,"slug":14,"sources":1963,"stem":1970,"suitable":16,"tagline":1971,"tags":1972,"updated":545,"verdict":1977,"website":1978,"__hash__":1979},"tools\u002Ftools\u002Fcoding\u002Flocal\u002Fcherry-studio.md",[15,541,12,13],{"type":18,"value":1541,"toc":1917},[1542,1544,1547,1550,1552,1596,1598,1610,1615,1619,1623,1640,1644,1661,1663,1687,1689,1813,1815,1847,1849,1872,1874,1892,1894],[21,1543,24],{"id":23},[26,1545,1546],{},"Cherry Studio 是一款开源、跨平台（Windows \u002F macOS \u002F Linux \u002F Android）的桌面 AI 客户端，定位『全能 AI 工作台』：把 OpenAI \u002F Anthropic \u002F Google \u002F DeepSeek 等云端模型，以及 Ollama \u002F LM Studio 本地模型，全部聚合到同一个桌面应用里管理。内置 300+ 助手模板、本地 RAG 知识库、Markdown + Mermaid 渲染、MCP 协议支持，所有对话数据本地存储 + WebDAV 备份。AGPL-3.0 开源、GitHub 60k+ stars，企业版可联系商务做私有化部署。",[26,1548,1549],{},"适合：中文 AI 重度用户、想统一管理多家模型、需要本地知识库 RAG、关注数据本地存储的开发者 \u002F 研究者。不适合：要 Web 端访问 \u002F Docker 自托管 \u002F 团队多人共享 \u002F iOS 端使用。",[21,1551,39],{"id":39},[41,1553,1554,1560,1566,1572,1578,1584,1590],{},[44,1555,1556,1559],{},[47,1557,1558],{},"多模型聚合","：OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek \u002F Qwen \u002F Kimi \u002F Moonshot 等云端 + Ollama \u002F LM Studio 本地",[44,1561,1562,1565],{},[47,1563,1564],{},"本地 RAG 知识库","：拖拽 PDF \u002F Word \u002F Excel \u002F PPT \u002F 网址 \u002F sitemap → 自动向量化 → 检索增强问答 + 来源追溯",[44,1567,1568,1571],{},[47,1569,1570],{},"300+ 助手模板","：编程 \u002F 写作 \u002F 翻译 \u002F 学习 \u002F 角色扮演开箱即用，可自定义 System Prompt",[44,1573,1574,1577],{},[47,1575,1576],{},"MCP 协议","：扩展工具调用 \u002F 联网搜索 \u002F 文件操作",[44,1579,1580,1583],{},[47,1581,1582],{},"数据本地优先","：对话历史本地存储，WebDAV 同步，不上传第三方",[44,1585,1586,1589],{},[47,1587,1588],{},"多模态","：图片识别 \u002F PDF 阅读 \u002F Markdown + Mermaid + 代码高亮",[44,1591,1592,1595],{},[47,1593,1594],{},"AI 绘画 + 翻译","：内置主流 SD \u002F DALL·E \u002F 翻译 API 集成",[21,1597,102],{"id":102},[41,1599,1600,1605],{},[44,1601,1602,1604],{},[47,1603,1025],{},"：完全免费，AGPL-3.0",[44,1606,1607,1609],{},[47,1608,1511],{},"：私有化部署 + 团队协作 + 资源管控，联系销售",[118,1611,1612],{},[26,1613,1614],{},"模型 API 费用按你自己绑定的供应商计费；本地 Ollama \u002F LM Studio 零成本。",[21,1616,1618],{"id":1617},"实测mac-m2-中型知识库","实测（Mac M2 + 中型知识库）",[26,1620,1621],{},[47,1622,131],{},[41,1624,1625,1628,1631,1634,1637],{},[44,1626,1627],{},"中文 UI \u002F 文档 \u002F 社区都顶级，零门槛上手",[44,1629,1630],{},"本地 RAG 拖入 30+ PDF 后向量化 \u003C 2 分钟（用 bge-m3）",[44,1632,1633],{},"多模型并排回答：让 Claude \u002F GPT \u002F DeepSeek 同回一个问题做比较",[44,1635,1636],{},"MCP 接 Brave Search + 自定义工具流畅",[44,1638,1639],{},"WebDAV 同步坚果云 \u002F 阿里云盘，桌面 + 移动设备数据互通",[26,1641,1642],{},[47,1643,153],{},[41,1645,1646,1649,1652,1655,1658],{},[44,1647,1648],{},"没有 Web 端 \u002F Docker 自托管（要这个用 LobeChat）",[44,1650,1651],{},"iOS 版尚未发布（roadmap 中）",[44,1653,1654],{},"大型 PDF（>100 MB）向量化偶有失败，要切小",[44,1656,1657],{},"助手市场质量参差，要自筛",[44,1659,1660],{},"模型 API 调用全靠你自己付费，新手要先理解 API Key 概念",[21,1662,173],{"id":173},[175,1664,1665,1668,1671,1678,1681,1684],{},[44,1666,1667],{},"cherry-ai.com 下载客户端（或 GitHub releases）",[44,1669,1670],{},"设置 → 模型服务 → 填 OpenAI \u002F Claude \u002F DeepSeek API Key",[44,1672,1673,1674],{},"（可选）本地：装 Ollama → Cherry Studio 自动识别 endpoint ",[428,1675,1676],{"href":1676,"rel":1677},"http:\u002F\u002Flocalhost:11434",[462],[44,1679,1680],{},"新建知识库 → 拖文件 \u002F 加网址 → 等向量化",[44,1682,1683],{},"新对话 → 选模型 → 勾知识库 → 提问",[44,1685,1686],{},"进阶：自定义助手（System Prompt）+ MCP 扩展工具",[21,1688,203],{"id":203},[205,1690,1691,1705],{},[208,1692,1693],{},[211,1694,1695,1697,1699,1701,1703],{},[214,1696,216],{},[214,1698,1250],{},[214,1700,1247],{},[214,1702,10],{},[214,1704,224],{},[229,1706,1707,1719,1733,1747,1760,1774,1789,1801],{},[211,1708,1709,1711,1713,1715,1717],{},[234,1710,236],{},[234,1712,1267],{},[234,1714,1264],{},[234,1716,1267],{},[234,1718,1261],{},[211,1720,1721,1723,1726,1728,1731],{},[234,1722,1558],{},[234,1724,1725],{},"✅ 云 + 本地",[234,1727,1725],{},[234,1729,1730],{},"本地为主",[234,1732,1725],{},[211,1734,1735,1738,1741,1743,1745],{},[234,1736,1737],{},"知识库 RAG",[234,1739,1740],{},"✅ 强",[234,1742,1740],{},[234,1744,319],{},[234,1746,273],{},[211,1748,1749,1752,1754,1756,1758],{},[234,1750,1751],{},"MCP",[234,1753,273],{},[234,1755,273],{},[234,1757,319],{},[234,1759,273],{},[211,1761,1762,1765,1768,1770,1772],{},[234,1763,1764],{},"自托管 \u002F Web",[234,1766,1767],{},"无 Web",[234,1769,1322],{},[234,1771,262],{},[234,1773,1322],{},[211,1775,1776,1779,1782,1784,1787],{},[234,1777,1778],{},"中文",[234,1780,1781],{},"5\u002F5",[234,1783,1781],{},[234,1785,1786],{},"4\u002F5",[234,1788,1786],{},[211,1790,1791,1793,1795,1797,1799],{},[234,1792,1347],{},[234,1794,1354],{},[234,1796,337],{},[234,1798,334],{},[234,1800,337],{},[211,1802,1803,1805,1807,1809,1811],{},[234,1804,1331],{},[234,1806,1340],{},[234,1808,1337],{},[234,1810,309],{},[234,1812,1334],{},[21,1814,360],{"id":360},[41,1816,1817,1823,1829,1835,1841],{},[44,1818,1819,1822],{},[47,1820,1821],{},"API Key 别明文外泄","：客户端配置文件以明文存 Key，机器借出前先清；团队共享用企业版 \u002F 自建中转",[44,1824,1825,1828],{},[47,1826,1827],{},"知识库别一次塞太多","：单库 1000+ 文档检索质量明显下降，按主题切分多个知识库",[44,1830,1831,1834],{},[47,1832,1833],{},"嵌入模型选择","：免费 bge-m3 够用；专业用付费 Pro\u002FBAAI\u002Fbge-m3 或 OpenAI text-embedding-3",[44,1836,1837,1840],{},[47,1838,1839],{},"WebDAV 同步先小范围测","：知识库向量数据较大，先备份对话再开同步",[44,1842,1843,1846],{},[47,1844,1845],{},"MCP 工具来源要可控","：MCP 是给 AI 真实工具能力，第三方插件审一遍代码",[21,1848,396],{"id":395},[41,1850,1851,1854,1857,1860,1863,1866,1869],{},[44,1852,1853],{},"✅ 中文用户、AI 重度使用 \u002F 多模型管理",[44,1855,1856],{},"✅ 需要本地 RAG 知识库",[44,1858,1859],{},"✅ 关注数据隐私 \u002F 本地存储",[44,1861,1862],{},"✅ 想用 Ollama \u002F LM Studio 本地模型",[44,1864,1865],{},"❌ 需要 Web 端 \u002F Docker 自托管",[44,1867,1868],{},"❌ 团队多人共享 \u002F SSO",[44,1870,1871],{},"❌ iOS 主力用户",[21,1873,422],{"id":422},[41,1875,1876,1880,1884,1888],{},[44,1877,1878],{},[428,1879,1436],{"href":1435},[44,1881,1882],{},[428,1883,953],{"href":516},[44,1885,1886],{},[428,1887,431],{"href":430},[44,1889,1890],{},[428,1891,967],{"href":966},[21,1893,452],{"id":452},[175,1895,1896,1903,1910],{},[44,1897,1898,1899],{},"Cherry Studio 官网（功能 + 下载）",[428,1900,1901],{"href":1901,"rel":1902},"https:\u002F\u002Fwww.cherry-ai.com\u002F",[462],[44,1904,1905,1906],{},"MBLUO Studio — Cherry Studio 评测 2026 ",[428,1907,1908],{"href":1908,"rel":1909},"https:\u002F\u002Fmbluostudio.com\u002Ftools\u002Fcherry-studio",[462],[44,1911,1912,1913],{},"Cursor IDE 博客 — Cherry Studio 完全指南（2025-03）",[428,1914,1915],{"href":1915,"rel":1916},"https:\u002F\u002Fwww.cursor-ide.com\u002Fblog\u002Fcherry-studio-guide",[462],{"title":478,"searchDepth":479,"depth":479,"links":1918},[1919,1920,1921,1922,1923,1924,1925,1926,1927,1928],{"id":23,"depth":482,"text":24},{"id":39,"depth":482,"text":39},{"id":102,"depth":482,"text":102},{"id":1617,"depth":482,"text":1618},{"id":173,"depth":482,"text":173},{"id":203,"depth":482,"text":203},{"id":360,"depth":482,"text":360},{"id":395,"depth":482,"text":396},{"id":422,"depth":482,"text":422},{"id":452,"depth":482,"text":452},"\u002Fimg\u002Ftools\u002Fcherry-studio.webp","Cherry Studio 真实评测：开源跨平台桌面 AI 客户端，集成 OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek + Ollama \u002F LM Studio 本地模型，内置 300+ 助手模板 + 本地 RAG 知识库。AGPL-3.0 开源、GitHub 60k+ stars，企业版另询。",[1932,1935,1938,1941],{"q":1933,"a":1934},"Cherry Studio 真的免费吗？","是。客户端完全免费、AGPL-3.0 开源，模型调用走你自己的 API Key（OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek 等付费）或本地 Ollama \u002F LM Studio（零成本）。",{"q":1936,"a":1937},"本地知识库怎么用？","在『知识库』面板新建，拖文件 \u002F 加网址 \u002F 填 sitemap，系统自动向量化（默认 BAAI\u002Fbge-m3 或硅基流动的 Pro 版）；提问时勾选要检索的知识库，AI 会基于检索片段答题并标出来源。",{"q":1939,"a":1940},"和 LobeChat 怎么选？","都开源、多模型、有 RAG。LobeChat 是 Web + 桌面双形态，可自托管 Docker，72k stars；Cherry Studio 是纯桌面（Win\u002FMac\u002FLinux\u002FAndroid），不支持 Web 部署但桌面体验更精细，60k+ stars。要 Web 访问 \u002F 公司多人共享选 LobeChat；个人重度选 Cherry Studio。",{"q":1942,"a":1943},"支持 MCP \u002F 插件吗？","支持 MCP（Model Context Protocol）扩展，配合自定义助手（System Prompt）可扩展工具调用、联网搜索等能力。","https:\u002F\u002Fgithub.com\u002FCherryHQ\u002Fcherry-studio",[513,512],{},[519,520,521,1948],"android",[1950,1953],{"plan":1025,"price":524,"features":1951,"notes":1952},"300+ 助手模板 \u002F 云端 + 本地模型 \u002F 知识库 \u002F MCP \u002F WebDAV 备份","AGPL-3.0 开源",{"plan":1511,"price":1954,"features":1955,"notes":1956},"联系销售","私有化部署 \u002F 团队协作 \u002F AI 资源管控 \u002F 知识库管理","面向企业团队","开源免费 \u002F 企业版联系销售",[534,1959],"onboarding\u002Fcursor-mcp-deep-integration",{"power":537,"ux":538,"price":538,"cn_support":538,"stability":537},{"title":1250,"description":1930},"Cherry Studio 评测 2026：AI 客户端工具，多模型桌面助手，开源免费",[1964,1966,1968],{"name":1965,"url":1901,"accessed":545},"Cherry Studio 官网",{"name":1967,"url":1908,"accessed":545},"MBLUO Studio — Cherry Studio 评测",{"name":1969,"url":1915,"accessed":545},"Cursor IDE 博客 — Cherry Studio 指南","tools\u002Fcoding\u002Flocal\u002Fcherry-studio","全能 AI 客户端：多模型聚合 + 本地知识库 + 300+ 助手模板，跨平台桌面应用",[492,1973,1974,1975,1531,1046,1976],"desktop","multi-model","knowledge-base","china","国产 AI 桌面客户端第一梯队，多模型聚合 + 本地 RAG + 中文体验顶级。需要 Web 部署 \u002F 自托管选 LobeChat；只要桌面体验完整选 Cherry Studio。","https:\u002F\u002Fcherry-ai.com","nhG3iaQ6G7dAWmMarVHUYX3EU47pSjb4yoUqlqZy18k",{"id":1981,"title":1247,"alternatives":1982,"api_compatible":1983,"body":1984,"category":492,"chinese_friendly":538,"cover":2417,"description":2418,"domestic":510,"extension":496,"faq":2419,"free":510,"github":2389,"languages":2432,"lastVerified":514,"meta":2433,"models":16,"navigation":510,"notSuitable":16,"opensource":510,"path":1435,"pillar":517,"platforms":2434,"priceTable":2436,"pricing":2443,"published":532,"relatedPlaybooks":2444,"relatedReviews":16,"score":2445,"self_host":510,"seo":2446,"seoTitle":2447,"slug":15,"sources":2448,"stem":2455,"suitable":16,"tagline":2456,"tags":2457,"updated":545,"verdict":2460,"website":2461,"__hash__":2462},"tools\u002Ftools\u002Fcoding\u002Flocal\u002Flobe-chat.md",[14,13,12,541],[567,568,569,570,571,572,573,574,575,576,577,578,579,221,580],{"type":18,"value":1985,"toc":2405},[1986,1988,1995,1998,2000,2062,2064,2077,2080,2084,2088,2108,2112,2129,2131,2157,2159,2292,2294,2332,2334,2360,2362,2380,2382],[21,1987,24],{"id":23},[26,1989,1990,1991,1994],{},"LobeChat 是 LobeHub 团队的开源 AI 聊天框架，2023 年发布、GitHub 72k+ stars、MIT 协议。",[47,1992,1993],{},"Web + 桌面 + Docker 自托管三形态","，把 OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek \u002F Qwen \u002F Kimi \u002F Ollama \u002F LM Studio 等 80+ 模型聚合到一个现代设计的客户端里。内置 RAG 知识库 + 插件市场 + 助手市场 + 多模型对比 + MCP，是当下综合最强的多模型 AI 客户端之一。",[26,1996,1997],{},"适合：需要 Web 端访问、Docker 自托管、多模型对比、丰富助手市场的用户；中文重度用户；想给团队 \u002F 家庭部署一个共享 AI 工作台。不适合：只用桌面 + 不需要 Web（Cherry Studio 同样优秀且更精细）、强企业 RBAC + 多租户（Open WebUI 多用户更完善）。",[21,1999,39],{"id":39},[41,2001,2002,2007,2013,2018,2024,2030,2035,2040,2046,2052],{},[44,2003,2004,2006],{},[47,2005,1558],{},"：OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek \u002F Qwen \u002F Kimi \u002F 豆包 \u002F Groq \u002F Together \u002F OpenRouter \u002F Ollama \u002F LM Studio",[44,2008,2009,2012],{},[47,2010,2011],{},"多模型对比","：同 prompt 给多模型并排回答",[44,2014,2015,2017],{},[47,2016,1564],{},"：上传 PDF \u002F Word \u002F 网页 → 向量化 → 检索引用",[44,2019,2020,2023],{},[47,2021,2022],{},"插件市场","：联网搜索 \u002F 代码执行 \u002F 图像生成 \u002F 翻译等几十款官方插件",[44,2025,2026,2029],{},[47,2027,2028],{},"助手市场","：几百个预设 AI 角色，一键导入",[44,2031,2032,2034],{},[47,2033,1576],{},"：扩展任意工具能力",[44,2036,2037],{},[47,2038,2039],{},"代码解释器 \u002F 文件上传 \u002F TTS \u002F 多模态",[44,2041,2042,2045],{},[47,2043,2044],{},"Web + 桌面 + Docker","：三形态，数据可完全本地",[44,2047,2048,2051],{},[47,2049,2050],{},"LobeHub Cloud","：官方云托管，免部署",[44,2053,2054,2057,2058,2061],{},[47,2055,2056],{},"快捷指令 \u002F 工作流","：自定义 prompt 模板，",[30,2059,2060],{},"\u002Fpodcast-summary"," 类用法",[21,2063,102],{"id":102},[41,2065,2066,2072],{},[44,2067,2068,2071],{},[47,2069,2070],{},"自托管 \u002F 桌面","：完全免费、MIT 开源",[44,2073,2074,2076],{},[47,2075,2050],{},"：订阅制，云端托管 + 团队协作 + 同步",[26,2078,2079],{},"模型 API 费用按你自己的供应商付费；本地 Ollama \u002F LM Studio 零成本。",[21,2081,2083],{"id":2082},"实测m2-自托管-docker连-openai-deepseek-本地-ollama","实测（M2 + 自托管 Docker，连 OpenAI + DeepSeek + 本地 Ollama）",[26,2085,2086],{},[47,2087,131],{},[41,2089,2090,2093,2096,2099,2102,2105],{},[44,2091,2092],{},"界面颜值是这一类工具里第一档（深色 \u002F 透明 \u002F 现代感）",[44,2094,2095],{},"多模型并排对比对选型极其有用：写一道复杂题，Claude \u002F GPT \u002F DeepSeek 直接对比答案",[44,2097,2098],{},"知识库 RAG 上传 50+ PDF 后检索准确，引用片段可视化",[44,2100,2101],{},"助手市场拿来即用——「Code Reviewer」「Translation Polish」节省 prompt 编写",[44,2103,2104],{},"Docker 一键部署，团队 5 人共享流畅",[44,2106,2107],{},"多平台数据同步（Cloud \u002F WebDAV）",[26,2109,2110],{},[47,2111,153],{},[41,2113,2114,2117,2120,2123,2126],{},[44,2115,2116],{},"自托管要熟悉 Docker + 反代 + HTTPS",[44,2118,2119],{},"国内连 OpenAI \u002F Claude 需自带网络方案",[44,2121,2122],{},"Web 版数据存 LobeHub，隐私敏感场景走桌面 \u002F Docker",[44,2124,2125],{},"插件市场质量参差，要自筛",[44,2127,2128],{},"团队多人共享需配 LobeHub Cloud 或自建数据库（Postgres + S3）",[21,2130,173],{"id":173},[175,2132,2133,2136,2142,2145,2148,2151,2154],{},[44,2134,2135],{},"选形态：Web（chat.lobehub.com 注册即用） \u002F 桌面（GitHub Releases 下载） \u002F Docker",[44,2137,2138,2139],{},"Docker：",[30,2140,2141],{},"docker run -d -p 3210:3210 -e OPENAI_API_KEY=sk-xxx --name lobe-chat lobehub\u002Flobe-chat",[44,2143,2144],{},"设置 → AI 服务商 → 添加 OpenAI \u002F Claude \u002F DeepSeek \u002F Ollama",[44,2146,2147],{},"模型选择器测试对话",[44,2149,2150],{},"知识库：拖文件 → 等向量化 → 对话引用",[44,2152,2153],{},"助手市场拉「Code Reviewer」「论文翻译润色」试用",[44,2155,2156],{},"进阶：插件市场启用联网搜索 \u002F 代码执行；MCP 自定义工具",[21,2158,203],{"id":203},[205,2160,2161,2175],{},[208,2162,2163],{},[211,2164,2165,2167,2169,2171,2173],{},[214,2166,216],{},[214,2168,1247],{},[214,2170,1250],{},[214,2172,224],{},[214,2174,10],{},[229,2176,2177,2189,2202,2214,2227,2243,2255,2268,2280],{},[211,2178,2179,2181,2183,2185,2187],{},[234,2180,236],{},[234,2182,2044],{},[234,2184,1267],{},[234,2186,1261],{},[234,2188,1267],{},[211,2190,2191,2193,2196,2198,2200],{},[234,2192,1558],{},[234,2194,2195],{},"✅ 80+",[234,2197,273],{},[234,2199,273],{},[234,2201,1730],{},[211,2203,2204,2206,2208,2210,2212],{},[234,2205,2011],{},[234,2207,1277],{},[234,2209,273],{},[234,2211,319],{},[234,2213,319],{},[211,2215,2216,2218,2220,2222,2225],{},[234,2217,1737],{},[234,2219,273],{},[234,2221,273],{},[234,2223,2224],{},"✅ + oikb",[234,2226,319],{},[211,2228,2229,2232,2235,2238,2241],{},[234,2230,2231],{},"插件 \u002F 助手市场",[234,2233,2234],{},"✅ 丰富",[234,2236,2237],{},"300+ 助手",[234,2239,2240],{},"Tools",[234,2242,319],{},[211,2244,2245,2247,2249,2251,2253],{},[234,2246,1751],{},[234,2248,273],{},[234,2250,273],{},[234,2252,1305],{},[234,2254,319],{},[211,2256,2257,2259,2262,2264,2266],{},[234,2258,1274],{},[234,2260,2261],{},"配 Cloud \u002F 自建",[234,2263,262],{},[234,2265,1277],{},[234,2267,262],{},[211,2269,2270,2272,2274,2276,2278],{},[234,2271,1331],{},[234,2273,1337],{},[234,2275,1340],{},[234,2277,1334],{},[234,2279,309],{},[211,2281,2282,2284,2286,2288,2290],{},[234,2283,1347],{},[234,2285,337],{},[234,2287,1354],{},[234,2289,337],{},[234,2291,334],{},[21,2293,360],{"id":360},[41,2295,2296,2302,2308,2314,2320,2326],{},[44,2297,2298,2301],{},[47,2299,2300],{},"Web 版数据不本地","：隐私敏感选桌面或 Docker 自托管",[44,2303,2304,2307],{},[47,2305,2306],{},"国内连海外模型走中转","：直连 OpenAI \u002F Claude 不稳，配 OpenRouter \u002F Ofox \u002F 国内中转",[44,2309,2310,2313],{},[47,2311,2312],{},"Docker 自托管暴露公网","：上反代 + HTTPS + Auth + 备份数据库",[44,2315,2316,2319],{},[47,2317,2318],{},"嵌入模型中文优化","：默认嵌入对中文一般，配 bge-m3 \u002F 硅基流动 Pro 版",[44,2321,2322,2325],{},[47,2323,2324],{},"插件市场审一遍","：第三方插件可执行代码，团队部署谨慎启用",[44,2327,2328,2331],{},[47,2329,2330],{},"同步选 Cloud vs WebDAV","：团队多端走 LobeHub Cloud；个人多设备 WebDAV 即可",[21,2333,396],{"id":395},[41,2335,2336,2339,2342,2345,2348,2351,2354,2357],{},[44,2337,2338],{},"✅ Web + 桌面双形态需求",[44,2340,2341],{},"✅ Docker 自托管 \u002F 团队共享",[44,2343,2344],{},"✅ 多模型对比 \u002F 选型",[44,2346,2347],{},"✅ 中文重度用户",[44,2349,2350],{},"✅ 助手市场 \u002F 插件生态用户",[44,2352,2353],{},"❌ 强企业 RBAC + 多租户（Open WebUI 更完善）",[44,2355,2356],{},"❌ 只要桌面 + 数据完全本地（Cherry Studio 同样优秀）",[44,2358,2359],{},"❌ 完全不会碰 Docker",[21,2361,422],{"id":422},[41,2363,2364,2368,2372,2376],{},[44,2365,2366],{},[428,2367,443],{"href":442},[44,2369,2370],{},[428,2371,437],{"href":436},[44,2373,2374],{},[428,2375,431],{"href":430},[44,2377,2378],{},[428,2379,967],{"href":966},[21,2381,452],{"id":452},[175,2383,2384,2391,2398],{},[44,2385,2386,2387],{},"LobeChat GitHub 仓库（72k+ stars，MIT）",[428,2388,2389],{"href":2389,"rel":2390},"https:\u002F\u002Fgithub.com\u002Flobehub\u002Flobe-chat",[462],[44,2392,2393,2394],{},"腾讯云开发者社区 — Lobe Chat 本地化 AI 聊天终极桌面客户端（2026-01）",[428,2395,2396],{"href":2396,"rel":2397},"https:\u002F\u002Fcloud.tencent.com\u002Fdeveloper\u002Farticle\u002F2622150",[462],[44,2399,2400,2401],{},"Ofox.ai — LobeChat 完全配置指南 2026（2026-04-17）",[428,2402,2403],{"href":2403,"rel":2404},"https:\u002F\u002Fofox.ai\u002Fzh\u002Fblog\u002Flobechat-api-configuration-guide-2026",[462],{"title":478,"searchDepth":479,"depth":479,"links":2406},[2407,2408,2409,2410,2411,2412,2413,2414,2415,2416],{"id":23,"depth":482,"text":24},{"id":39,"depth":482,"text":39},{"id":102,"depth":482,"text":102},{"id":2082,"depth":482,"text":2083},{"id":173,"depth":482,"text":173},{"id":203,"depth":482,"text":203},{"id":360,"depth":482,"text":360},{"id":395,"depth":482,"text":396},{"id":422,"depth":482,"text":422},{"id":452,"depth":482,"text":452},"\u002Fimg\u002Ftools\u002Flobe-chat.webp","LobeChat 真实评测：LobeHub 团队开源 AI 聊天框架，GitHub 72k+ stars、MIT 协议。Web + 桌面（Win\u002FMac\u002FLinux\u002FDocker）双形态，支持 OpenAI \u002F Claude \u002F Gemini \u002F DeepSeek \u002F Qwen \u002F Ollama 等 80+ 模型，内置 RAG 知识库 + 插件市场 + 助手市场 + 多模型对比。",[2420,2423,2426,2429],{"q":2421,"a":2422},"Web 版 vs 桌面版 vs Docker 自托管，怎么选？","Web 版（chat.lobehub.com）最快上手但数据存 LobeHub 服务器；桌面版数据本地存、隐私好；Docker 自托管对团队 \u002F 公司部署最优，完全掌控数据。",{"q":2424,"a":2425},"支持哪些模型？","80+ 模型：OpenAI 全系列、Anthropic Claude、Google Gemini、DeepSeek、Qwen、Kimi、Moonshot、字节豆包、Groq、Together、OpenRouter、Ollama \u002F LM Studio 本地模型，以及任何 OpenAI 兼容 API。",{"q":2427,"a":2428},"多模型对比怎么用？","同一对话窗口里把消息广播给多个模型并排回答，选型 \u002F 评估特别有用——直接看 Claude 和 GPT 在同一 prompt 下的回答差异。",{"q":2430,"a":2431},"助手市场是什么？","LobeHub 维护的预设 AI 角色市场（代码审查 \u002F 翻译 \u002F 写作 \u002F 角色扮演等几百个），一键拉到本地用，省去自己写 System Prompt。",[513,512],{},[2435,519,520,521,1022],"web",[2437,2439],{"plan":2070,"price":524,"features":2438,"notes":1509},"全功能 \u002F 80+ 模型 \u002F 知识库 \u002F 插件 \u002F 助手市场",{"plan":2050,"price":2440,"features":2441,"notes":2442},"订阅制","云端托管 \u002F 免部署 \u002F 团队协作 \u002F 同步","chat.lobehub.com 注册即用","完全免费（MIT 开源） \u002F LobeHub Cloud 订阅",[534,535],{"power":538,"ux":538,"price":538,"cn_support":538,"stability":537},{"title":1247,"description":2418},"LobeChat - 开源 AI 聊天框架评测与部署 | AIHO",[2449,2451,2453],{"name":2450,"url":2389,"accessed":545},"LobeChat GitHub",{"name":2452,"url":2396,"accessed":545},"腾讯云开发者社区 — Lobe Chat 终极桌面客户端",{"name":2454,"url":2403,"accessed":545},"Ofox.ai — LobeChat 完全配置指南 2026","tools\u002Fcoding\u002Flocal\u002Flobe-chat","现代设计的开源 AI 聊天框架——Web + 桌面双形态、72k+ stars、多模型 + 知识库 + 插件市场",[492,2435,1973,1974,1531,2458,2459,1046],"plugin","mcp","颜值与功能双优的多模型 AI 聊天客户端。要 Web + 桌面双形态、自托管 Docker、多模型对比、丰富助手市场——LobeChat 是综合最强；纯桌面体验 Cherry Studio 同样优秀。","https:\u002F\u002Flobehub.com","sAhQ5CHDobJa--z6NbtgysTQDwTBYsoGPXR0zat-5-w",1785660641109]