{"id":15,"date":"2026-06-01T09:08:15","date_gmt":"2026-06-01T01:08:15","guid":{"rendered":"http:\/\/tantan.ink\/?p=15"},"modified":"2026-06-01T09:08:15","modified_gmt":"2026-06-01T01:08:15","slug":"openjarvis-v1-local-ai","status":"publish","type":"post","link":"https:\/\/www.tantan.ink\/index.php\/2026\/06\/01\/openjarvis-v1-local-ai\/","title":{"rendered":"OpenJarvis \u53d1\u5e03 1.0\uff1a\u628a\u4e2a\u4eba AI \u52a9\u624b\u771f\u6b63\u8dd1\u5728\u81ea\u5df1\u786c\u4ef6\u4e0a"},"content":{"rendered":"<p>\u6700\u8fd1\u770b\u5230 Stanford Hazy Research \u548c Scaling Intelligence \u5b9e\u9a8c\u5ba4\u53d1\u5e03\u4e86 OpenJarvis \u7684 1.0 \u7248\u672c\u3002\u8fd9\u662f\u4e00\u4e2a\u5f00\u6e90\u6846\u67b6\uff0c\u7528\u6765\u5728\u672c\u5730\u786c\u4ef6\u4e0a\u8dd1\u4e2a\u4eba AI \u52a9\u624b\uff0c\u5e76\u4e14\u5185\u7f6e\u4e86\u5bf9 Ollama \u7684\u652f\u6301\u3002<\/p>\n<p>\u7c7b\u4f3c\u7684&#8221;\u4e2a\u4eba AI \u52a9\u624b&#8221;\u6982\u5ff5\u6700\u8fd1\u51fa\u4e86\u4e0d\u5c11\uff0c\u4f46 OpenJarvis \u6709\u4e00\u4e2a\u8bbe\u8ba1\u601d\u8def\u8ba9\u6211\u89c9\u5f97\u6bd4\u8f83\u5bf9\u8def\uff1a\u300clocal-first\uff0c\u4e91\u7aef\u662f\u53ef\u9009\u9879\u300d\uff0c\u800c\u4e0d\u662f\u53cd\u8fc7\u6765\u3002<\/p>\n<h2>\u5b83\u548c\u522b\u7684\u6709\u4ec0\u4e48\u533a\u522b\uff1f<\/h2>\n<p>\u5f53\u524d\u5927\u591a\u6570\u4e2a\u4eba AI \u4ea7\u54c1\u9ed8\u8ba4\u628a\u6240\u6709\u8bf7\u6c42\u53d1\u5230\u4e91\u7aef\u3002OpenJarvis \u7684\u51fa\u53d1\u70b9\u4e0d\u540c\uff1a\u6a21\u578b\u5728\u672c\u5730\u8dd1\uff0c\u4e91\u7aef\u53ea\u5728\u9700\u8981\u65f6\u624d\u4ecb\u5165\u3002\u5b83\u51fa\u81ea Stanford \u7684&#8221;Intelligence Per Watt&#8221;\u7814\u7a76\u65b9\u5411\uff0c\u5173\u6ce8\u7684\u4e0d\u53ea\u662f\u6a21\u578b\u7cbe\u5ea6\uff0c\u8fd8\u5305\u62ec\u80fd\u8017\u3001\u5ef6\u8fdf\u3001\u6210\u672c\u8fd9\u4e9b\u5b9e\u9645\u843d\u5730\u65f6\u4f1a\u78b0\u5230\u7684\u6307\u6807\u3002<\/p>\n<p>\u5bf9\u6298\u817e\u672c\u5730 AI \u7684\u4eba\u6765\u8bf4\uff0c\u8fd9\u51e0\u4e2a\u70b9\u6bd4\u8f83\u5b9e\u9645\uff1a<\/p>\n<ul>\n<li><strong>\u4e0d\u5f3a\u5236\u8054\u7f51<\/strong>\u3002\u65e5\u5e38\u804a\u5929\u3001\u95ee\u7b54\u3001\u6587\u4ef6\u641c\u7d22\u5168\u5728\u672c\u5730\u5b8c\u6210\u3002\u7f51\u7edc\u4e0d\u662f\u524d\u7f6e\u6761\u4ef6\u3002<\/li>\n<li><strong>\u548c Ollama \u6df1\u5ea6\u6253\u901a<\/strong>\u3002\u5b89\u88c5\u811a\u672c\u4f1a\u81ea\u52a8\u68c0\u6d4b\u5df2\u6709\u7684 Ollama \u5b89\u88c5\uff0c\u4e0d\u7528\u624b\u52a8\u914d API \u5730\u5740\u3002<\/li>\n<li><strong>\u5185\u7f6e\u9884\u8bbe agent<\/strong>\u3002\u4e0d\u9700\u8981\u4ece\u96f6\u642d agent \u6846\u67b6\uff0c\u5b83\u76f4\u63a5\u5e26\u4e86\u51e0\u4e2a\u5f00\u7bb1\u5373\u7528\u7684\u9884\u8bbe\uff1a\u65e9\u6668\u7b80\u62a5\u3001\u6df1\u5ea6\u7814\u7a76\u3001\u672c\u5730\u7f16\u7a0b\u52a9\u624b\u3002<\/li>\n<\/ul>\n<h2>\u4e0a\u624b\u5927\u6982\u4ec0\u4e48\u6837<\/h2>\n<p>\u524d\u63d0\u662f\u5df2\u7ecf\u88c5\u4e86 Ollama\u3002\u7136\u540e\u4e00\u884c\u547d\u4ee4\u5b89\u88c5\uff1a<\/p>\n<pre><code>curl -fsSL https:\/\/open-jarvis.github.io\/OpenJarvis\/install.sh | bash<\/code><\/pre>\n<p>\u5b89\u88c5\u5b8c\u6210\u540e <code>jarvis<\/code> \u547d\u4ee4\u5c31\u53ef\u4ee5\u7528\u4e86\u3002\u5b89\u88c5\u811a\u672c\u4f1a\u62c9\u4e00\u4e2a\u5165\u95e8\u6a21\u578b\u8ba9\u4f60\u76f4\u63a5\u8dd1\u8d77\u6765\uff0c\u4e5f\u53ef\u4ee5\u81ea\u5df1\u6307\u5b9a\uff1a<\/p>\n<pre><code>jarvis model pull qwen3.5:35b\njarvis ask -m qwen3.5:35b \"\u4f60\u7684\u95ee\u9898\"<\/code><\/pre>\n<p>\u9ed8\u8ba4\u6a21\u578b\u5199\u5230 <code>~\/.openjarvis\/config.toml<\/code> \u5c31\u884c\uff1a<\/p>\n<pre><code>[intelligence]\ndefault_model = \"qwen3.5:35b\"\npreferred_engine = \"ollama\"<\/code><\/pre>\n<p>Windows \u7528\u6237\u5728 WSL2 \u91cc\u8dd1\uff0c\u6216\u8005\u76f4\u63a5\u88c5\u684c\u9762\u7248\u3002<\/p>\n<h2>\u4e09\u4e2a\u9884\u8bbe agent \u80fd\u505a\u4ec0\u4e48<\/h2>\n<p>\u5b89\u88c5\u5b8c\u5e76\u4e0d\u662f\u4e00\u4e2a\u7a7a\u6846\u67b6\u3002<code>jarvis init --preset<\/code> \u53ef\u4ee5\u9009\u62e9\u9884\u8bbe\uff0c\u6bcf\u4e2a\u9884\u8bbe\u628a\u5f15\u64ce\u548c\u5de5\u5177\u90fd\u6253\u5305\u597d\u4e86\uff1a<\/p>\n<p><strong>\u65e9\u6668\u7b80\u62a5\uff08morning-digest-mac\uff09<\/strong>\uff1a\u8bfb\u53d6\u65e5\u5386\u3001\u90ae\u4ef6\u548c\u5f53\u5929\u65b0\u95fb\uff0c\u751f\u6210\u4e00\u4efd\u7b80\u62a5\u3002\u9002\u5408\u6bcf\u5929\u65e9\u4e0a\u8dd1\u4e00\u4e0b\u3002<\/p>\n<p><strong>\u6df1\u5ea6\u7814\u7a76\uff08deep-research\uff09<\/strong>\uff1a\u53ef\u4ee5\u7d22\u5f15\u672c\u5730\u6587\u4ef6\u76ee\u5f55\uff0c\u7136\u540e\u7528\u81ea\u7136\u8bed\u8a00\u8de8\u6587\u4ef6\u548c\u7f51\u7edc\u505a\u641c\u7d22\uff0c\u8fd4\u56de\u5e26\u5f15\u7528\u7684\u56de\u7b54\u3002<\/p>\n<p><strong>\u672c\u5730\u7f16\u7a0b\u52a9\u624b\uff08code-assistant\uff09<\/strong>\uff1a\u5728\u672c\u5730\u5199 Python \u4ee3\u7801\u5e76\u6267\u884c\u3002\u76f8\u5f53\u4e8e\u4e00\u4e2a\u672c\u5730\u4ee3\u7801 agent\uff0c\u4e0d\u7528\u628a\u4ee3\u7801\u4f20\u5230\u4efb\u4f55\u4e91\u7aef\u670d\u52a1\u3002<\/p>\n<p>\u4e3e\u4e2a\u5b9e\u9645\u4f8b\u5b50\uff0c\u5047\u8bbe\u4f60\u60f3\u6574\u7406\u9879\u76ee\u6587\u6863\uff1a<\/p>\n<pre><code>jarvis init --preset deep-research\njarvis memory index .\/docs\/\njarvis ask \"\u603b\u7ed3\u6240\u6709\u548c Project X \u76f8\u5173\u7684\u90ae\u4ef6\"<\/code><\/pre>\n<h2>\u503c\u5f97\u5173\u6ce8\u4f46\u522b\u671f\u671b\u592a\u9ad8\u7684\u4e8b<\/h2>\n<ol>\n<li><strong>\u8fd9\u662f 1.0 \u7248\u672c<\/strong>\u3002\u6846\u67b6\u521a\u53d1\u5e03\uff0c\u751f\u6001\u8fd8\u5728\u65e9\u671f\u3002\u9884\u8bbe agent \u5728\u5f53\u524d\u9636\u6bb5\u66f4\u591a\u662f&#8221;\u80fd\u8dd1\u901a&#8221;\u800c\u975e&#8221;\u6210\u719f\u7a33\u5b9a&#8221;\u3002\u6df1\u5ea6\u7814\u7a76\u90a3\u4e2a\u9884\u8bbe\u5bf9\u6587\u4ef6\u91cf\u5927\u7684\u573a\u666f\u53ef\u80fd\u6bd4\u8f83\u6162\u3002<\/li>\n<li><strong>\u672c\u5730\u6a21\u578b\u80fd\u529b\u51b3\u5b9a\u4e0a\u9650<\/strong>\u3002OpenJarvis \u672c\u8eab\u662f\u8c03\u5ea6\u5c42\uff0c\u771f\u6b63\u5e72\u6d3b\u7684\u662f\u5e95\u4e0b\u7684\u6a21\u578b\u3002\u8dd1 Qwen 3.5 35B \u548c\u8dd1\u4e00\u4e2a 7B \u5c0f\u6a21\u578b\uff0cagent \u7684\u8868\u73b0\u4f1a\u6709\u672c\u8d28\u533a\u522b\u3002<\/li>\n<li><strong>\u4f9d\u8d56 Ollama \u4f5c\u4e3a\u63a8\u7406\u5f15\u64ce<\/strong>\u3002\u5982\u679c\u4f60\u5df2\u7ecf\u5728\u7528 Ollama\uff0c\u96c6\u6210\u4f53\u9a8c\u4f1a\u5f88\u987a\uff1b\u5982\u679c\u4f60\u7528\u7684\u662f LM Studio \u6216\u5176\u4ed6\u63a8\u7406\u540e\u7aef\uff0c\u76ee\u524d\u8fd8\u4e0d\u652f\u6301\u3002<\/li>\n<li><strong>macOS\/Linux \u4f18\u5148<\/strong>\u3002Windows \u8981\u8d70 WSL2 \u8def\u5f84\uff0c\u6ca1\u6709\u539f\u751f Windows \u652f\u6301\u3002<\/li>\n<\/ol>\n<h2>\u6211\u7684\u5224\u65ad<\/h2>\n<p>OpenJarvis \u662f\u90a3\u79cd&#8221;\u65b9\u5411\u5bf9\u4e86\u4f46\u8fd8\u9700\u8981\u65f6\u95f4&#8221;\u7684\u9879\u76ee\u3002\u5b83\u7684\u4ef7\u503c\u4e0d\u5728\u4e8e\u529f\u80fd\u591a\u5168\uff0c\u800c\u5728\u4e8e\u628a local-first \u8fd9\u4e2a\u7406\u5ff5\u505a\u6210\u4e86\u80fd\u8dd1\u7684\u4e1c\u897f\uff0c\u5e76\u4e14\u7ed9\u4e86\u51e0\u4e2a\u770b\u5f97\u89c1\u6478\u5f97\u7740\u7684 agent \u9884\u8bbe\u3002\u5bf9\u5df2\u7ecf\u5728\u8dd1 Ollama \u7684\u4eba\uff0c\u82b1\u534a\u5c0f\u65f6\u88c5\u4e0a\u8bd5\u8bd5\u4e0d\u4f1a\u6709\u635f\u5931\u3002\u4f46\u5982\u679c\u671f\u671b\u5b83\u9a6c\u4e0a\u66ff\u4ee3\u73b0\u5728\u7684 AI \u5de5\u4f5c\u6d41\uff0c\u53ef\u80fd\u8981\u5931\u671b\u2014\u20141.0 \u66f4\u591a\u662f\u9a8c\u8bc1\u8def\u7ebf\uff0c\u4e0d\u662f\u6253\u5305\u6240\u6709\u529f\u80fd\u3002<\/p>\n<p>\u6211\u4f1a\u7ee7\u7eed\u5173\u6ce8\u5b83\u7684\u66f4\u65b0\u3002\u5982\u679c\u540e\u9762\u6df1\u5ea6\u7814\u7a76\u90a3\u4e2a\u9884\u8bbe\u53d8\u5f97\u597d\u7528\uff0c\u5199\u6587\u7ae0\u548c\u7814\u7a76\u7c7b\u4efb\u52a1\u4e5f\u8bb8\u80fd\u7701\u4e0d\u5c11\u4e8b\u3002<\/p>\n<hr>\n<p><em>\u6765\u6e90\uff1aOllama Blog \u2014 <a href=\"https:\/\/ollama.com\/blog\/openjarvis\">OpenJarvis: a local-first personal AI is now available to run with Ollama<\/a>\uff082026-05-28\uff09<\/em><br \/>\n<em>GitHub\uff1a<a href=\"https:\/\/github.com\/open-jarvis\/OpenJarvis\">https:\/\/github.com\/open-jarvis\/OpenJarvis<\/a><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Stanford \u53d1\u5e03 OpenJarvis 1.0\uff0c\u4e00\u4e2a\u672c\u5730\u4f18\u5148\u7684\u4e2a\u4eba AI \u6846\u67b6\uff0c\u5185\u7f6e Ollama \u652f\u6301\u3002\u672c\u6587\u4ecb\u7ecd\u4e0a\u624b\u65b9\u6cd5\u548c\u4e09\u4e2a\u9884\u8bbe agent\u3002<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[],"tags":[],"class_list":["post-15","post","type-post","status-publish","format-standard","hentry"],"_links":{"self":[{"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/posts\/15","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/comments?post=15"}],"version-history":[{"count":0,"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/posts\/15\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/media?parent=15"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/categories?post=15"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.tantan.ink\/index.php\/wp-json\/wp\/v2\/tags?post=15"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}