https://virgoc0der.github.io/tags/ollama/Recent content in Ollama on Billy's BlogHugozh-CNbillychen826@gmail.com (Billy)billychen826@gmail.com (Billy)Mon, 24 Mar 2025 15:56:48 +0800https://virgoc0der.github.io/posts/first/Mon, 24 Mar 2025 15:56:48 +0800billychen826@gmail.com (Billy)https://virgoc0der.github.io/posts/first/<h2 id="环境准备">环境准备</h2> <ol> <li>系统要求： <ul> <li>macOS Monterey 12.3 或更高版本</li> <li>至少16GB内存（推荐32GB）</li> <li>安装Homebrew包管理器</li> </ul> </li> </ol> <div class="highlight"><pre tabindex="0" style="color:#c0caf5;background-color:#1a1b26;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-bash" data-lang="bash"><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 安装Homebrew</span> </span></span><span style="display:flex;"><span>/bin/bash -c <span style="color:#9ece6a">&#34;</span><span style="color:#bb9af7">$(</span>curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh<span style="color:#bb9af7">)</span><span style="color:#9ece6a">&#34;</span> </span></span></code></pre></div><h2 id="安装部署">安装部署</h2> <h3 id="1-安装ollama">1. 安装Ollama</h3> <div class="highlight"><pre tabindex="0" style="color:#c0caf5;background-color:#1a1b26;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-bash" data-lang="bash"><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 通过curl直接安装</span> </span></span><span style="display:flex;"><span>curl -fsSL https://ollama.com/install.sh | sh </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 启动ollama服务</span> </span></span><span style="display:flex;"><span>ollama serve </span></span></code></pre></div><h3 id="2-部署open-webui">2. 部署Open WebUI</h3> <div class="highlight"><pre tabindex="0" style="color:#c0caf5;background-color:#1a1b26;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-bash" data-lang="bash"><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 使用Docker运行webui容器</span> </span></span><span style="display:flex;"><span>docker run -d -p 3000:8080 <span style="color:#7aa2f7">\ </span></span></span><span style="display:flex;"><span><span style="color:#7aa2f7"></span> -v ollama:/root/.ollama <span style="color:#7aa2f7">\ </span></span></span><span style="display:flex;"><span><span style="color:#7aa2f7"></span> --name open-webui <span style="color:#7aa2f7">\ </span></span></span><span style="display:flex;"><span><span style="color:#7aa2f7"></span> ghcr.io/open-webui/open-webui:main </span></span></code></pre></div><h2 id="模型加载">模型加载</h2> <div class="highlight"><pre tabindex="0" style="color:#c0caf5;background-color:#1a1b26;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-bash" data-lang="bash"><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 下载llama3模型（根据需求替换模型名称）</span> </span></span><span style="display:flex;"><span>ollama pull llama3 </span></span><span style="display:flex;"><span> </span></span><span style="display:flex;"><span><span style="color:#414868;font-style:italic"># 查看已安装模型</span> </span></span><span style="display:flex;"><span>ollama list </span></span></code></pre></div><h2 id="访问验证">访问验证</h2> <ol> <li> <p>打开浏览器访问：</p> <ul> <li>Ollama管理界面：http://localhost:11434</li> <li>WebUI界面：http://localhost:3000</li> </ul> </li> <li> <p>在WebUI界面选择模型即可开始对话</p> </li> </ol> <h2 id="注意事项">注意事项</h2> <ol> <li>确保3000/11434端口未被占用</li> <li>首次下载模型需要较长时间（取决于网络环境）</li> <li>建议配合Ngrok实现内网穿透</li> <li>使用GPU加速需要额外配置Metal后端</li> </ol>