[{"data":1,"prerenderedAt":193},["ShallowReactive",2],{"repo-qiao-920-llama-cpp-desktop-zh-CN":3,"i18n-qiao-920-llama-cpp-desktop-zh-CN":190},{"mirrorPool":4,"detail":13,"relatedProjects":75},[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":26,"homepage":27,"pushedAt":28,"ownerAvatar":29,"ownerName":30,"ownerBio":31,"lastSyncTime":32,"lastReleaseTime":33,"tags":34,"license":35,"category":36,"subgroup":37,"releases":38},"Qiao-920","llama-cpp-desktop",1217873104,"https:\u002F\u002Fgithub.com\u002FQiao-920\u002Fllama-cpp-desktop",385,"JavaScript","Windows desktop control panel for local llama.cpp server",{"JavaScript":22,"CSS":23,"PowerShell":24,"HTML":25},80.6,18.1,1.2,0.1,[],"","2026-07-13","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F239728274?v=4","Qiaofeng","I have done data analysis, video editing and three-dimensional special effects. Now I want to be a developer.Come on, come on, I will succeed","2026-10-04T13:17:41.339897+00:00","2026-07-13T14:14:05Z",[],"MIT","ai","LLM推理\u002F微调工具",[39,53,64],{"tag":40,"publishAt":33,"semVer":41,"prerelease":45,"assets":46},"v0.6.14",[42,43,44],0,6,14,false,[47],{"fileName":48,"sizeMb":49,"sha256":50,"downloadCount":51,"originUrl":52},"Llama.cpp-Desktop.exe",86.66,"00cefae3b73c08d94ec8722c22a5994d6249bef29caf79096e973afd978d5000",3363,"https:\u002F\u002Fgithub.com\u002FQiao-920\u002Fllama-cpp-desktop\u002Freleases\u002Fdownload\u002Fv0.6.14\u002FLlama.cpp-Desktop.exe",{"tag":54,"publishAt":55,"semVer":56,"prerelease":45,"assets":58},"v0.6.13","2026-05-04T15:21:36Z",[42,43,57],13,[59],{"fileName":48,"sizeMb":60,"sha256":61,"downloadCount":62,"originUrl":63},86.64,"946abd522fc7df61e84c46009b95e8ffce900c946a917f3d5cd06aeba25a7320",6156,"https:\u002F\u002Fgithub.com\u002FQiao-920\u002Fllama-cpp-desktop\u002Freleases\u002Fdownload\u002Fv0.6.13\u002FLlama.cpp-Desktop.exe",{"tag":65,"publishAt":66,"semVer":67,"prerelease":45,"assets":69},"v0.6.12","2026-04-22T10:16:18Z",[42,43,68],12,[70],{"fileName":48,"sizeMb":71,"sha256":72,"downloadCount":73,"originUrl":74},86.63,"87287e72846aa753cd9b84cfa00ddd272c18df6b374f6c8dba7bdf31c9948d01",457,"https:\u002F\u002Fgithub.com\u002FQiao-920\u002Fllama-cpp-desktop\u002Freleases\u002Fdownload\u002Fv0.6.12\u002FLlama.cpp-Desktop.exe",[76,94,127,147,159,176],{"owner":77,"repo":78,"repoId":79,"category":36,"subgroup":37,"language":80,"tags":81,"starCount":83,"ownerUid":84,"latestTag":85,"platforms":86,"lastReleaseTime":91,"downloadCount":92,"description":93},"ggml-org","llama.cpp",612354784,"C++",[82],"ggml",131695,134263123,"b11734",[87,88,89,90],"windows","mac","linux","other","2026-10-19T14:08:40Z",481,"一个用C\u002FC++实现的大语言模型（LLM）推理引擎，支持多种模型格式，提供CLI、服务器和内置Web UI，可在本地运行LLM和VLM。",{"owner":95,"repo":96,"repoId":97,"category":36,"subgroup":37,"language":98,"tags":99,"starCount":120,"ownerUid":121,"latestTag":122,"platforms":123,"lastReleaseTime":124,"downloadCount":125,"description":126},"diegosouzapw","OmniRoute",1157102282,"TypeScript",[100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119],"a2a","ai-agents","ai-gateway","anthropic","claude","claude-code","cline","codex","copilot","cursor","deepseek","free-ai","gemini","kimi","llm-gateway","mcp","openai","openai-proxy","qwen","token-saver",72815,8016841,"v3.8.51",[87,88,89,90],"2026-09-30T01:41:50Z",4681,"一款免费的 MIT AI 网关，提供统一端点访问 290+ 提供商、500+ 模型，支持智能回退和令牌压缩。",{"owner":128,"repo":129,"repoId":130,"category":131,"subgroup":37,"language":132,"tags":133,"starCount":140,"ownerUid":141,"latestTag":142,"platforms":143,"lastReleaseTime":144,"downloadCount":145,"description":146},"AlexsJones","llmfit",1158588224,"tools","Rust",[134,135,136,137,138,139],"gguf","llm","localai","mlx","skill","unsloth",37743,1235925,"v1.1.17",[87,88,89],"2026-10-08T18:04:05Z",1447,"一个帮助用户根据硬件配置推荐最佳本地大语言模型的命令行工具，支持多GPU、MoE架构和多种量化格式。",{"owner":148,"repo":149,"repoId":150,"category":36,"subgroup":37,"language":98,"tags":151,"starCount":152,"ownerUid":153,"latestTag":154,"platforms":155,"lastReleaseTime":156,"downloadCount":157,"description":158},"tashfeenahmed","freellmapi",1217040185,[],37411,9307356,"v0.16.5",[87,88,89,90],"2026-10-19T11:30:09Z",1332,"一个聚合了34个免费LLM提供商、635个免费模型端点的统一API网关，每月提供约74亿免费令牌，支持智能路由、自动故障转移和加密密钥存储。",{"owner":160,"repo":161,"repoId":162,"category":36,"subgroup":37,"language":163,"tags":164,"starCount":169,"ownerUid":170,"latestTag":171,"platforms":172,"lastReleaseTime":173,"downloadCount":174,"description":175},"jundot","omlx",1157171418,"Python",[165,166,135,167,137,168],"apple-silicon","inference-server","macos","openai-api",22858,64250138,"v0.7.1rc1",[88,90],"2026-10-15T18:23:10Z",2195,"一个专为Apple Silicon优化的LLM推理服务器，支持连续批处理和SSD缓存，通过菜单栏进行管理。",{"owner":177,"repo":178,"repoId":179,"category":36,"subgroup":37,"language":180,"tags":181,"starCount":183,"ownerUid":184,"latestTag":185,"platforms":186,"lastReleaseTime":187,"downloadCount":188,"description":189},"yetone","magpie",1383809012,"Go",[105,107,110,182,135,167],"gemini-cli",16805,1206493,"v0.1.1293",[87,88],"2026-10-19T12:31:24Z",309,"Magpie 是一个本地运行的 AI 网关，让 Claude Code、Codex、Gemini CLI、OpenCode 等编程代理统一使用 Kimi、DeepSeek、GLM、ChatGPT 等模型，菜单栏一键切换，配额用尽自动切换下一账号。",{"description":191,"readmeHtml":192},"Llama.cpp Desktop 是一款为 Windows 本地 llama.cpp 服务器设计的桌面控制面板，将启动服务、配置模型、查看日志、直接聊天和接入 OpenAI 兼容客户端整合在一个窗口中，让用户无需命令行即可管理本地 llama.cpp 推理服务。","\u003Ch2 id=\"项目概述\">项目概述\u003C\u002Fh2>\u003Cp>Llama.cpp Desktop 是一个给 Windows 本地 llama.cpp 准备的桌面端控制面板。启动服务、配置模型、查看日志、直接聊天和接入 OpenAI 兼容客户端，都放在一个窗口里。\u003C\u002Fp>\n\u003Ch2 id=\"主要功能\">主要功能\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>本地直连\u003C\u002Fstrong>：直接启动 llama.cpp 原版目录里的 llama-server.exe，不强依赖额外启动器\u003C\u002Fli>\n\u003Cli>\u003Cstrong>OpenAI 兼容\u003C\u002Fstrong>：默认提供 \u003Ca href=\"http:\u002F\u002F127.0.0.1:8080\u002Fv1，可接入\" target=\"_blank\" rel=\"noopener noreferrer\">http:\u002F\u002F127.0.0.1:8080\u002Fv1，可接入\u003C\u002Fa> OpenClaw、Claude Code 等客户端\u003C\u002Fli>\n\u003Cli>\u003Cstrong>桌面聊天\u003C\u002Fstrong>：内置网页端风格聊天页面，支持流式回复、历史对话、搜索和消息操作\u003C\u002Fli>\n\u003Cli>\u003Cstrong>附件入口\u003C\u002Fstrong>：支持图片、文本、PDF 等附件入口，图片可在聊天里预览\u003C\u002Fli>\n\u003Cli>\u003Cstrong>模型信息\u003C\u002Fstrong>：点击模型标签即可查看当前模型、上下文、GPU 层数和运行参数\u003C\u002Fli>\n\u003Cli>\u003Cstrong>终端日志\u003C\u002Fstrong>：在客户端里查看 llama.cpp 输出，方便排查启动和推理问题\u003C\u002Fli>\n\u003Cli>\u003Cstrong>托盘后台\u003C\u002Fstrong>：关闭窗口后隐藏到系统托盘，服务继续后台运行\u003C\u002Fli>\n\u003Cli>\u003Cstrong>参数配置\u003C\u002Fstrong>：支持模型路径、上下文、采样、GPU 层数、线程和批处理参数\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"适用人群\">适用人群\u003C\u002Fh2>\u003Cul>\n\u003Cli>Windows 10\u002F11 用户\u003C\u002Fli>\n\u003Cli>需要管理本地 llama.cpp 推理服务的开发者\u003C\u002Fli>\n\u003Cli>希望以图形化界面替代命令行的用户\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"快速上手\">快速上手\u003C\u002Fh2>\u003Col>\n\u003Cli>从 Releases 下载 Llama.cpp-Desktop.exe\u003C\u002Fli>\n\u003Cli>双击运行\u003C\u002Fli>\n\u003Cli>在设置里选择 llama.cpp 原文件目录，或直接选择 llama-server.exe\u003C\u002Fli>\n\u003Cli>选择 GGUF 模型文件\u003C\u002Fli>\n\u003Cli>保存配置并启动服务，使用内置聊天或接入 OpenAI 兼容客户端\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Ch2 id=\"核心优势\">核心优势\u003C\u002Fh2>\u003Cul>\n\u003Cli>图形化界面，无需记忆复杂命令行参数\u003C\u002Fli>\n\u003Cli>内置聊天页面，开箱即用\u003C\u002Fli>\n\u003Cli>支持托盘后台运行，不干扰其他工作\u003C\u002Fli>\n\u003Cli>可接入 OpenClaw、Claude Code 等外部客户端\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"系统要求\">系统要求\u003C\u002Fh2>\u003Cul>\n\u003Cli>操作系统：Windows 10 \u002F 11\u003C\u002Fli>\n\u003Cli>额外依赖：需本机已有可用的 llama.cpp Windows 构建目录和 GGUF 模型文件；项目不内置 llama.cpp、模型文件、显卡驱动或 CUDA\u002FVulkan 运行库\u003C\u002Fli>\n\u003C\u002Ful>\n",1792450963949]