[{"data":1,"prerenderedAt":206},["ShallowReactive",2],{"repo-ddalcu-mlx-serve":3,"i18n-ddalcu-mlx-serve-zh-CN":203},{"mirrorPool":4,"detail":13},[5,9],{"key":6,"label":7,"base":8,"type":6},"github","mirror.github","https:\u002F\u002Fgithub.com\u002F",{"key":10,"label":11,"base":12,"type":10},"auto-proxy","mirror.autoProxy","https:\u002F\u002Fgithub.xxlab.tech\u002F",{"owner":14,"repo":15,"repoId":16,"repoUrl":17,"starCount":18,"language":19,"description":20,"languages":21,"screenshots":32,"homepage":33,"pushedAt":34,"ownerAvatar":35,"ownerName":36,"ownerBio":37,"lastSyncTime":38,"lastReleaseTime":39,"tags":40,"license":61,"category":62,"subgroup":63,"releases":64},"ddalcu","mlx-serve",1159656889,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve",1644,"Zig","Native LLM inference server for Apple Silicon. OpenAI + Anthropic API compatible. No Python. Includes MLX Core macOS app with chat, agent mode, and tool calling.",{"Zig":22,"Swift":23,"Shell":24,"C++":24,"Python":25,"HTML":26,"JavaScript":27,"C":28,"CSS":29,"Objective-C":29,"Jinja":30,"Metal":30,"Dockerfile":31,"Ruby":31,"Makefile":31},54.2,24.6,6.3,4.1,2.4,1.1,0.4,0.2,0.1,0,[],"http:\u002F\u002Fmlxserve.com\u002F","2026-09-28","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F869085?v=4","David Dalcu","I like to build things.","2026-09-28T15:20:53.755410+00:00","2026-09-26T05:24:09Z",[41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60],"agent","anthropic-api","apple-silicon","claude-code","deepseek-v4","diffusion","gguf","image-generation","inference","llm","local-llm","macos","macos-app","mlx","openai-api","tool-calling","video-generation","voice-agent","voice-cloning","zig","NOASSERTION","ai","本地LLM服务",[65,85,107,123,139,155,170,186],{"tag":66,"publishAt":39,"semVer":67,"prerelease":71,"assets":72},"v26.9.6",[68,69,70],26,9,6,false,[73,79],{"fileName":74,"sizeMb":75,"sha256":76,"downloadCount":77,"originUrl":78},"mlx-serve-bin-macos-arm64.tar.gz",69.02,"bb4ec3f6ee250745a1d04ef685d3f15b734b560f482a65d0e8bacb512e78bacf",775,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.6\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":80,"sizeMb":81,"sha256":82,"downloadCount":83,"originUrl":84},"MLX-Serve.dmg",82.42,"e149f4d0acdfa1312fb2ca084141e4fb878c046bb8651bafbb32736559f71269",1521,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.6\u002FMLX-Serve.dmg",{"tag":86,"publishAt":87,"semVer":88,"prerelease":71,"assets":90},"v26.9.5","2026-09-21T15:26:40Z",[68,69,89],5,[91,96,101],{"fileName":74,"sizeMb":92,"sha256":93,"downloadCount":94,"originUrl":95},68.9,"06c087e623a729070fb4b51832bf6acbafc3037ac23368b469f25066413cdbba",996,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.5\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":80,"sizeMb":97,"sha256":98,"downloadCount":99,"originUrl":100},82.43,"772354e9e6c3915b29609ff8f07a8837f69273853371a2e385144c31a23689af",10,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.5\u002FMLX-Serve.dmg",{"fileName":102,"sizeMb":103,"sha256":104,"downloadCount":105,"originUrl":106},"MLXCore.dmg",81.85,"b1e7c5e97b61db3e577129a625e6b0c66d1581ffbc9b6c09814c8c949d6c5035",2405,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.5\u002FMLXCore.dmg",{"tag":108,"publishAt":109,"semVer":110,"prerelease":111,"assets":112},"v26.9.5-pre-release.1","2026-09-18T17:44:01Z",[68,69,89],true,[113,118],{"fileName":74,"sizeMb":114,"sha256":115,"downloadCount":116,"originUrl":117},68.81,"64639234eb9021107e7a76837356ff13929c837c7f8d57ce61dee06ac933343a",86,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.5-pre-release.1\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":119,"sha256":120,"downloadCount":121,"originUrl":122},81.81,"2f742ed3d6dcb4dafa9de214c1e56af1c9425677c203b86d38f3c119b35a7b9f",108,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.5-pre-release.1\u002FMLXCore.dmg",{"tag":124,"publishAt":125,"semVer":126,"prerelease":71,"assets":128},"v26.9.4","2026-09-17T07:21:00Z",[68,69,127],4,[129,134],{"fileName":74,"sizeMb":130,"sha256":131,"downloadCount":132,"originUrl":133},68.74,"5a8b16317e7e4d5f87f528384a5d28b70bacbdcfce291be2aa6513a87e2a1925",477,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.4\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":135,"sha256":136,"downloadCount":137,"originUrl":138},81.67,"116b356badb447dfdd9b371ae0b3f2940908b0f83495803b58bde36b182d6032",1753,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.4\u002FMLXCore.dmg",{"tag":140,"publishAt":141,"semVer":142,"prerelease":71,"assets":144},"v26.9.3","2026-09-16T06:05:40Z",[68,69,143],3,[145,150],{"fileName":74,"sizeMb":146,"sha256":147,"downloadCount":148,"originUrl":149},68.73,"05863d67dc823798ac072ef72824360cbbd868734d2f79516bba153349eef794",255,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.3\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":151,"sha256":152,"downloadCount":153,"originUrl":154},81.28,"14f698936b089200875aba0252429deee7bc327c11cfb5f8bbd204990fffc142",753,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.3\u002FMLXCore.dmg",{"tag":156,"publishAt":157,"semVer":158,"prerelease":71,"assets":160},"v26.9.2","2026-09-09T00:24:44Z",[68,69,159],2,[161,166],{"fileName":74,"sizeMb":162,"sha256":163,"downloadCount":164,"originUrl":165},67.74,"f2dd5f80d4e50e4ebd57b28e70ce79ed95a657337360784975b610cec50400af",2034,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.2\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":97,"sha256":167,"downloadCount":168,"originUrl":169},"4fa51eed22cfafe058ba20e9b50353cb1316ba44041382e1d20eb71d2a34680e",2874,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.2\u002FMLXCore.dmg",{"tag":171,"publishAt":172,"semVer":173,"prerelease":71,"assets":175},"v26.9.1","2026-09-03T19:44:09Z",[68,69,174],1,[176,181],{"fileName":74,"sizeMb":177,"sha256":178,"downloadCount":179,"originUrl":180},67.52,"b9bb5178ac2dcfbfa232ffa1e6ce77e87a6408a540cbdaabb66d101d667590b0",768,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.1\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":182,"sha256":183,"downloadCount":184,"originUrl":185},82,"88a970efa61e422d9fbeb72fa1c8f4049e19d19d9252f2529ecbefe73e23bcc4",1782,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.9.1\u002FMLXCore.dmg",{"tag":187,"publishAt":188,"semVer":189,"prerelease":71,"assets":192},"v26.8.11","2026-08-29T14:03:23Z",[68,190,191],8,11,[193,198],{"fileName":74,"sizeMb":194,"sha256":195,"downloadCount":196,"originUrl":197},67.46,"8a6790d4540635f69301da99a64d81e7dad68fa43f3712c77b79fdd03187b3c3",936,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.8.11\u002Fmlx-serve-bin-macos-arm64.tar.gz",{"fileName":102,"sizeMb":199,"sha256":200,"downloadCount":201,"originUrl":202},81.89,"430c533cdf86909a87b54d7dc9431262a269bd73932f5f72aba220a731ccfd98",3015,"https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve\u002Freleases\u002Fdownload\u002Fv26.8.11\u002FMLXCore.dmg",{"description":204,"readmeHtml":205},"专为Apple Silicon优化的原生LLM推理服务器，支持OpenAI和Anthropic兼容API，无需Python，支持MLX和GGUF模型，附带macOS菜单栏应用。","\u003Ch2 id=\"项目概述\">项目概述\u003C\u002Fh2>\u003Cp>mlx-serve 是一个为Apple Silicon优化的原生LLM推理服务器，支持OpenAI和Anthropic兼容API。它运行任何MLX格式模型和HuggingFace上的GGUF模型（Qwen、Llama、Mistral、Gemma、DeepSeek等），支持图像、视频、音乐、语音（含语音克隆）和3D模型生成。\u003C\u002Fp>\n\u003Ch2 id=\"主要功能\">主要功能\u003C\u002Fh2>\u003Cul>\n\u003Cli>OpenAI和Anthropic兼容HTTP API\u003C\u002Fli>\n\u003Cli>支持MLX和GGUF模型\u003C\u002Fli>\n\u003Cli>支持图像、视频、音乐、语音（含语音克隆）和3D模型生成\u003C\u002Fli>\n\u003Cli>MLX Core macOS菜单栏应用：聊天、代理模式、MCP工具调用、模型管理\u003C\u002Fli>\n\u003Cli>Ollama API兼容（\u003Ccode>\u002Fapi\u002Fchat\u003C\u002Fcode>、\u003Ccode>\u002Fapi\u002Fgenerate\u003C\u002Fcode>等）\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"适用人群\">适用人群\u003C\u002Fh2>\u003Cul>\n\u003Cli>Apple Silicon Mac用户\u003C\u002Fli>\n\u003Cli>需要本地LLM推理的开发者\u003C\u002Fli>\n\u003Cli>希望替代LM Studio的用户\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"快速上手\">快速上手\u003C\u002Fh2>\u003Col>\n\u003Cli>\u003Cstrong>推荐\u003C\u002Fstrong>：下载MLX Core.app（签名公证的macOS菜单栏应用）\u003C\u002Fli>\n\u003Cli>Homebrew安装：\u003Ccode>brew tap ddalcu\u002Fmlx-serve https:\u002F\u002Fgithub.com\u002Fddalcu\u002Fmlx-serve && brew install --cask mlx-core\u003C\u002Fcode>\u003C\u002Fli>\n\u003Cli>CLI使用：\u003Ccode>mlx-serve run gemma4\u003C\u002Fcode>\u003C\u002Fli>\n\u003Cli>服务器地址：\u003Ccode>http:\u002F\u002Flocalhost:11234\u003C\u002Fcode>\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Ch2 id=\"核心优势\">核心优势\u003C\u002Fh2>\u003Cul>\n\u003Cli>专为Apple Silicon优化，比LM Studio更快\u003C\u002Fli>\n\u003Cli>无需Python，无需云端，无Electron\u003C\u002Fli>\n\u003Cli>支持OpenAI、Anthropic和Ollama三种API兼容\u003C\u002Fli>\n\u003Cli>开源免费\u003C\u002Fli>\n\u003C\u002Ful>\n",1790699343968]