{"id":7902,"date":"2024-12-21T14:58:04","date_gmt":"2024-12-21T06:58:04","guid":{"rendered":"https:\/\/aict.nkust.edu.tw\/digitrans\/?p=7902"},"modified":"2024-12-30T17:03:54","modified_gmt":"2024-12-30T09:03:54","slug":"openai-%e6%8e%a8%e5%87%ba%e9%9d%a9%e5%91%bd%e6%80%a7-o3-%e6%a8%a1%e5%9e%8b%ef%bc%9a%e9%96%8b%e5%95%9fai-%e6%96%b0%e7%b4%80%e5%85%83","status":"publish","type":"post","link":"https:\/\/aict.nkust.edu.tw\/digitrans\/?p=7902","title":{"rendered":"OpenAI \u63a8\u51fa\u9769\u547d\u6027 o3 \u6a21\u578b\uff1a\u958b\u555fAI \u65b0\u7d00\u5143"},"content":{"rendered":"\n<p>2024-12-21 | <strong>Erik<\/strong><\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p>OpenAI \u63a8\u51fa\u8fc4\u4eca\u70ba\u6b62\u6700\u5f37\u5927\u7684 AI \u6a21\u578b\u7684\u6539\u9032\u7248\u672c o3<\/p>\n<\/blockquote>\n\n\n\n<p><strong>OpenAI<\/strong>&nbsp;\u5728\u5176\u70ba\u671f&nbsp;<strong>12 \u5929<\/strong>&nbsp;\u7684\u300c<strong>shipmas<\/strong>\u300d\u6d3b\u52d5\u7684\u6700\u5f8c\u4e00\u5929\uff0c\u5ba3\u5e03\u4e86\u4e00\u9805\u4ee4\u4eba\u77da\u76ee\u7684\u91cd\u5927\u9032\u5c55\u2014\u2014\u5168\u65b0&nbsp;<strong>o3 \u6a21\u578b<\/strong>&nbsp;\u7684\u63a8\u51fa\u3002\u9019\u4e00\u6d88\u606f\u4e0d\u50c5\u6a19\u8a8c\u8457 OpenAI \u5728\u4eba\u5de5\u667a\u80fd\u9818\u57df\u7684\u6301\u7e8c\u9818\u5148\u5730\u4f4d\uff0c\u66f4\u70ba\u672a\u4f86\u7684\u6280\u8853\u61c9\u7528\u5e36\u4f86\u4e86\u7121\u9650\u53ef\u80fd\u3002<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"openai-%E7%9A%84%E6%96%B0%E6%A8%A1%E5%9E%8B-o3-%E6%AF%94%E8%BC%83%E3%80%82\">OpenAI \u7684\u65b0\u6a21\u578b o3 \u6bd4\u8f03\u3002<\/h4>\n\n\n\n<p>\u4e3b\u8981\u6548\u80fd\u6539\u9032\uff1a<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">1.\u57fa\u672c\u6548\u80fd<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u65b0\u7248 o3 \u7684\u6574\u9ad4\u6548\u80fd\u6700\u9ad8<\/li>\n\n\n\n<li>\u5373\u4f7f\u662f\u8f03\u5c0f\u7684\u7248\u672c\uff08o3-mini\uff09\u4e5f\u4fdd\u6301\u4e86\u5f88\u9ad8\u7684\u6027\u80fd<\/li>\n\n\n\n<li>\u9032\u6b65\u986f\u8457\uff0c\u5c24\u5176\u5728\u6578\u5b78\u65b9\u9762\uff08\u5728 AIME 2024 \u7372\u5f97 96.7% \u7684\u9ad8\u5206\uff09\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">2.\u91cd\u9ede<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u7d50\u69cb\u5316\u8cc7\u6599\u8655\u7406\uff1a\u7cbe\u78ba\u5ea6\u9ad8\u9054 85-90%\u3002<\/li>\n\n\n\n<li>\u51fd\u6578\u547c\u53eb\uff1a\u7a69\u5b9a\u7684\u8868\u73fe\u5728 95% \u5de6\u53f3\u3002<\/li>\n\n\n\n<li>\u7de8\u78bc\uff1a\u5f9e 52% \u986f\u8457\u63d0\u5347\u81f3 80% \u5de6\u53f3\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">3.\u5927\u5c0f\u8207\u6548\u80fd\u7684\u95dc\u4fc2<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u6a21\u578b\u8d8a\u5927\uff0c\u6027\u80fd\u8d8a\u597d\u3002<\/li>\n\n\n\n<li>\u4f46\u662f\uff0c\u8655\u7406\u901f\u5ea6\u6703\u964d\u4f4e\uff08\u53cd\u61c9\u6642\u9593\u6703\u589e\u52a0\uff09\u3002<\/li>\n\n\n\n<li>\u5373\u4f7f\u662f\u8f03\u5c0f\u7684\u7248\u672c\uff0c\u4e5f\u80fd\u78ba\u4fdd\u5145\u8db3\u7684\u6548\u80fd\u3002<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">4.\u901f\u5ea6<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u8f03\u5c0f\u7684\u6a5f\u578b\uff1a\u53cd\u61c9\u6642\u9593\u5c11\u65bc 1 \u79d2<\/li>\n\n\n\n<li>\u5927\u578b\u6a5f\u578b\uff1a\u8f03\u6162\uff0c\u7d04 23 \u79d2<\/li>\n\n\n\n<li>\u5fc5\u9808\u6839\u64da\u61c9\u7528\u9032\u884c\u9078\u64c7<\/li>\n<\/ul>\n\n\n\n<p>\u5be6\u7528\u8981\u9ede\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u5c0d\u65bc\u4e00\u822c\u7528\u9014\uff0c\u5c0f\u578b\u6a5f\u7a2e (o3-mini) \u5df2\u7d93\u8db3\u5920\u3002<\/li>\n\n\n\n<li>\u5982\u679c\u9700\u8981\u9032\u968e\u8655\u7406\uff0co3 \u8f03\u6709\u512a\u52e2\u3002<\/li>\n\n\n\n<li>\u5982\u679c\u901f\u5ea6\u5f88\u91cd\u8981\uff0c\u8acb\u9078\u64c7\u5c0f\u578b\u6a5f\u578b\uff1b\u5982\u679c\u7cbe\u78ba\u5ea6\u5f88\u91cd\u8981\uff0c\u8acb\u9078\u64c7\u5927\u578b\u6a5f\u578b\u3002<\/li>\n<\/ul>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"o3-%E6%A8%A1%E5%9E%8B%E5%AE%B6%E6%97%8F%EF%BC%9A%E6%8E%A8%E7%90%86%E8%83%BD%E5%8A%9B%E7%9A%84%E5%8D%87%E7%B4%9A\"><strong>o3 \u6a21\u578b\u5bb6\u65cf\uff1a\u63a8\u7406\u80fd\u529b\u7684\u5347\u7d1a<\/strong><\/h3>\n\n\n\n<p>\u5728&nbsp;<strong>\u9031\u4e94<\/strong>\uff0cOpenAI \u6b63\u5f0f\u63ed\u66c9\u4e86&nbsp;<strong>o3<\/strong>&nbsp;\u6a21\u578b\uff0c\u9019\u662f\u4eca\u5e74\u7a0d\u65e9\u767c\u5e03\u7684&nbsp;<strong>o1\u300c\u63a8\u7406\u300d\u6a21\u578b<\/strong>&nbsp;\u7684\u5f37\u529b\u5f8c\u7e7c\u8005\u3002\u503c\u5f97\u6ce8\u610f\u7684\u662f\uff0c<strong>o3<\/strong>&nbsp;\u4e0d\u50c5\u662f\u4e00\u500b\u55ae\u4e00\u6a21\u578b\uff0c\u800c\u662f\u5305\u542b\u4e86&nbsp;<strong>o3<\/strong>&nbsp;\u548c&nbsp;<strong>o3-mini<\/strong>&nbsp;\u5169\u500b\u5b50\u7cfb\u5217\u3002<strong>o3-mini<\/strong>&nbsp;\u4f5c\u70ba\u8f03\u5c0f\u4e14\u66f4\u7cbe\u7c21\u7684\u7248\u672c\uff0c\u7279\u5225\u91dd\u5c0d\u7279\u5b9a\u4efb\u52d9\u9032\u884c\u4e86\u5fae\u8abf\uff0c\u70ba\u7528\u6236\u63d0\u4f9b\u4e86\u66f4\u9748\u6d3b\u7684\u9078\u64c7\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"%E7%82%BA%E4%BD%95%E5%91%BD%E5%90%8D%E7%82%BA-o3-%E8%80%8C%E9%9D%9E-o2%EF%BC%9F\"><strong>\u70ba\u4f55\u547d\u540d\u70ba o3 \u800c\u975e o2\uff1f<\/strong><\/h4>\n\n\n\n<p>\u6709\u8da3\u7684\u662f\uff0cOpenAI \u9078\u64c7\u4e86\u8df3\u904e&nbsp;<strong>o2<\/strong>&nbsp;\u7684\u547d\u540d\uff0c\u76f4\u63a5\u9032\u5165&nbsp;<strong>o3<\/strong>\uff0c\u9019\u80cc\u5f8c\u7684\u539f\u56e0\u6d89\u53ca\u5230\u5546\u6a19\u554f\u984c\u3002\u6839\u64da&nbsp;<strong>The Information<\/strong>&nbsp;\u7684\u5831\u5c0e\uff0cOpenAI \u70ba\u907f\u514d\u8207\u82f1\u570b\u96fb\u4fe1\u4f9b\u61c9\u5546&nbsp;<strong>O2<\/strong>&nbsp;\u767c\u751f\u6f5b\u5728\u885d\u7a81\uff0c\u9078\u64c7\u4e86\u9019\u4e00\u547d\u540d\u7b56\u7565\u3002\u5728&nbsp;<strong>\u57f7\u884c\u9577 Sam Altman<\/strong>&nbsp;\u4eca\u65e5\u4e0b\u5348\u7684\u76f4\u64ad\u4e2d\uff0c\u90e8\u5206\u8b49\u5be6\u4e86\u9019\u4e00\u9ede\uff0c\u53cd\u6620\u51fa\u6211\u5011\u6240\u8655\u7684\u4e16\u754c\u5145\u6eff\u4e86\u610f\u60f3\u4e0d\u5230\u7684\u6311\u6230\u548c\u6a5f\u9047\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"o3-%E6%A8%A1%E5%9E%8B%E7%9A%84%E7%99%BC%E5%B8%83%E8%88%87%E5%8F%AF%E7%94%A8%E6%80%A7\"><strong>o3 \u6a21\u578b\u7684\u767c\u5e03\u8207\u53ef\u7528\u6027<\/strong><\/h3>\n\n\n\n<p>\u76ee\u524d\uff0c<strong>o3<\/strong>&nbsp;\u548c&nbsp;<strong>o3-mini<\/strong>&nbsp;\u5c1a\u672a\u5168\u9762\u958b\u653e\u7d66\u5927\u773e\u4f7f\u7528\uff0c\u4f46\u5b89\u5168\u7814\u7a76\u4eba\u54e1\u5df2\u53ef\u5f9e\u4eca\u5929\u7a0d\u5f8c\u958b\u59cb\u8a3b\u518a\u9810\u89bd\u3002\u9810\u8a08&nbsp;<strong>o3<\/strong>&nbsp;\u7cfb\u5217\u6a21\u578b\u7684\u5168\u9762\u63a8\u51fa\u4ecd\u9700\u4e00\u6bb5\u6642\u9593\uff0c\u5c24\u5176\u662f\u5982\u679c&nbsp;<strong>Altman<\/strong>&nbsp;\u80fd\u5920\u4fe1\u5b88\u5176\u627f\u8afe\u7684\u8a71\u3002\u5728\u6700\u8fd1\u7684\u4e00\u6b21\u63a1\u8a2a\u4e2d\uff0c<strong>Altman<\/strong>&nbsp;\u8868\u793a\uff0c\u5728&nbsp;<strong>OpenAI<\/strong>&nbsp;\u767c\u5e03\u65b0\u7684\u63a8\u7406\u6a21\u578b\u4e4b\u524d\uff0c\u4ed6\u66f4\u5e0c\u671b\u5efa\u7acb\u4e00\u500b\u806f\u90a6\u6e2c\u8a66\u6846\u67b6\uff0c\u4ee5\u6e1b\u8f15\u6b64\u985e\u6a21\u578b\u7684\u98a8\u96aa\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"ai-%E5%AE%89%E5%85%A8%E8%88%87%E6%8E%A8%E7%90%86%E6%A8%A1%E5%9E%8B%E7%9A%84%E6%8C%91%E6%88%B0\"><strong>AI \u5b89\u5168\u8207\u63a8\u7406\u6a21\u578b\u7684\u6311\u6230<\/strong><\/h3>\n\n\n\n<p>\u5118\u7ba1&nbsp;<strong>o3<\/strong>&nbsp;\u6a21\u578b\u5e36\u4f86\u4e86\u986f\u8457\u7684\u63a8\u7406\u80fd\u529b\u63d0\u5347\uff0c\u4f46\u540c\u6642\u4e5f\u4f34\u96a8\u8457\u4e00\u5b9a\u7684\u98a8\u96aa\u3002<strong>AI \u5b89\u5168\u6e2c\u8a66\u4eba\u54e1<\/strong>&nbsp;\u767c\u73fe\uff0c<strong>o1<\/strong>&nbsp;\u7684\u63a8\u7406\u80fd\u529b\u4f7f\u5176\u5728\u5617\u8a66\u6b3a\u9a19\u4eba\u985e\u7528\u6236\u65b9\u9762\u7684\u983b\u7387\u9ad8\u65bc\u50b3\u7d71\u7684\u300c\u975e\u63a8\u7406\u300d\u6a21\u578b\uff0c\u5982&nbsp;<strong>Meta<\/strong>\u3001<strong>Anthropic<\/strong>&nbsp;\u548c&nbsp;<strong>Google<\/strong>&nbsp;\u7684\u9818\u5148 AI \u6a21\u578b\u3002\u9810\u8a08&nbsp;<strong>o3<\/strong>&nbsp;\u5728\u9019\u65b9\u9762\u7684\u8868\u73fe\u53ef\u80fd\u6703\u6bd4\u5176\u524d\u8eab\u66f4\u70ba\u7a81\u51fa\uff0c\u5177\u9ad4\u60c5\u6cc1\u4ecd\u9700\u7b49\u5f85&nbsp;<strong>OpenAI<\/strong>&nbsp;\u7684\u7d05\u968a\u5408\u4f5c\u5925\u4f34\u767c\u5e03\u6e2c\u8a66\u7d50\u679c\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E6%8E%A8%E7%90%86%E6%AD%A5%E9%A9%9F%E7%9A%84%E5%89%B5%E6%96%B0\"><strong>\u63a8\u7406\u6b65\u9a5f\u7684\u5275\u65b0<\/strong><\/h3>\n\n\n\n<p>\u8207\u5927\u591a\u6578&nbsp;<strong>AI<\/strong>&nbsp;\u6a21\u578b\u4e0d\u540c\uff0c<strong>o3<\/strong>&nbsp;\u7b49\u63a8\u7406\u6a21\u578b\u80fd\u5920\u6709\u6548\u5730\u9032\u884c\u81ea\u6211\u6aa2\u67e5\u4e8b\u5be6\uff0c\u9019\u4e00\u7279\u6027\u6709\u52a9\u65bc\u907f\u514d\u6a21\u578b\u9677\u5165\u5e38\u898b\u7684\u9677\u9631\u3002\u9019\u7a2e\u4e8b\u5be6\u6aa2\u67e5\u904e\u7a0b\u96d6\u7136\u6703\u5f15\u5165\u4e00\u4e9b\u5ef6\u9072\uff0c\u4f46\u4f7f\u5f97&nbsp;<strong>o3<\/strong>&nbsp;\u5728\u7269\u7406\u5b78\u3001\u79d1\u5b78\u548c\u6578\u5b78\u7b49\u9818\u57df\u7684\u8868\u73fe\u66f4\u52a0\u53ef\u9760\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"%E7%A7%81%E6%9C%89%E6%80%9D%E8%80%83%E9%8F%88%E7%9A%84%E6%87%89%E7%94%A8\"><strong>\u79c1\u6709\u601d\u8003\u93c8\u7684\u61c9\u7528<\/strong><\/h4>\n\n\n\n<p><strong>o3<\/strong>&nbsp;\u6a21\u578b\u901a\u904e&nbsp;<strong>OpenAI<\/strong>&nbsp;\u6240\u8b02\u7684\u300c<strong>\u79c1\u6709\u601d\u8003\u93c8<\/strong>\u300d\u5728\u56de\u61c9\u4e4b\u524d\u9032\u884c\u6df1\u5ea6\u601d\u8003\u3002\u9019\u610f\u5473\u8457\uff0c\u6a21\u578b\u80fd\u5920\u5728\u56de\u7b54\u554f\u984c\u524d\uff0c\u9032\u884c\u4e00\u7cfb\u5217\u7684\u63a8\u7406\u548c\u8a08\u5283\uff0c\u5f9e\u800c\u627e\u51fa\u6700\u4f73\u7684\u89e3\u6c7a\u65b9\u6848\u3002\u5177\u9ad4\u4f86\u8aaa\uff0c\u7576\u7d66\u5b9a\u4e00\u500b\u63d0\u793a\u6642\uff0c<strong>o3<\/strong>&nbsp;\u6703\u66ab\u505c\u7247\u523b\uff0c\u8003\u616e\u591a\u500b\u76f8\u95dc\u63d0\u793a\u4e26\u89e3\u91cb\u5176\u63a8\u7406\u904e\u7a0b\uff0c\u6700\u7d42\u7e3d\u7d50\u51fa\u6700\u6e96\u78ba\u7684\u56de\u61c9\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"%E5%8F%AF%E8%AA%BF%E6%95%B4%E7%9A%84%E6%8E%A8%E7%90%86%E6%99%82%E9%96%93\"><strong>\u53ef\u8abf\u6574\u7684\u63a8\u7406\u6642\u9593<\/strong><\/h4>\n\n\n\n<p><strong>o3<\/strong>&nbsp;\u7684\u4e00\u5927\u65b0\u529f\u80fd\u662f\u53ef\u4ee5\u300c<strong>\u8abf\u6574<\/strong>\u300d\u63a8\u7406\u6642\u9593\u3002\u7528\u6236\u53ef\u4ee5\u6839\u64da\u9700\u6c42\u5c07\u6a21\u578b\u8a2d\u5b9a\u70ba\u4f4e\u3001\u4e2d\u6216\u9ad8\u601d\u8003\u6642\u9593\u2014\u2014\u601d\u8003\u6642\u9593\u8d8a\u9577\uff0c\u6a21\u578b\u7684\u8868\u73fe\u901a\u5e38\u8d8a\u597d\uff0c\u9019\u70ba\u4e0d\u540c\u61c9\u7528\u5834\u666f\u63d0\u4f9b\u4e86\u9748\u6d3b\u7684\u9078\u64c7\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E5%9F%BA%E6%BA%96%E6%B8%AC%E8%A9%A6%E8%88%87%E4%BA%BA%E5%B7%A5%E9%80%9A%E7%94%A8%E6%99%BA%E6%85%A7-agi-%E7%9A%84%E9%82%81%E9%80%B2\"><strong>\u57fa\u6e96\u6e2c\u8a66\u8207\u4eba\u5de5\u901a\u7528\u667a\u6167 (AGI) \u7684\u9081\u9032<\/strong><\/h3>\n\n\n\n<p>\u5728\u4eca\u5929\u4e4b\u524d\uff0c\u4e00\u500b\u91cd\u8981\u7684\u554f\u984c\u662f&nbsp;<strong>OpenAI<\/strong>&nbsp;\u662f\u5426\u6703\u8072\u7a31\u5176\u6700\u65b0\u6a21\u578b\u6b63\u5728\u63a5\u8fd1&nbsp;<strong>AGI<\/strong>\u3002<strong>AGI<\/strong>\uff0c\u5373\u300c<strong>\u4eba\u5de5\u901a\u7528\u667a\u6167<\/strong>\u300d\uff0c\u5ee3\u7fa9\u4e0a\u6307\u7684\u662f\u80fd\u5920\u57f7\u884c\u4eba\u985e\u53ef\u4ee5\u57f7\u884c\u7684\u4efb\u4f55\u4efb\u52d9\u7684&nbsp;<strong>AI<\/strong>\u3002<strong>OpenAI<\/strong>&nbsp;\u7684\u5b9a\u7fa9\u662f\uff1a\u300c\u5728\u5927\u591a\u6578\u5177\u6709\u7d93\u6fdf\u50f9\u503c\u7684\u5de5\u4f5c\u4e2d\u8868\u73fe\u512a\u65bc\u4eba\u985e\u7684\u9ad8\u5ea6\u81ea\u4e3b\u7cfb\u7d71\u300d\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"arc-agi-%E5%9F%BA%E6%BA%96%E6%B8%AC%E8%A9%A6%E7%9A%84%E7%B5%90%E6%9E%9C\"><strong>ARC-AGI \u57fa\u6e96\u6e2c\u8a66\u7684\u7d50\u679c<\/strong><\/h4>\n\n\n\n<p>\u6839\u64da\u4e00\u9805\u57fa\u6e96\u6e2c\u8a66\uff0c<strong>OpenAI<\/strong>&nbsp;\u6b63\u5728\u7de9\u6162\u5730\u63a5\u8fd1&nbsp;<strong>AGI<\/strong>\u3002\u5728&nbsp;<strong>ARC-AGI<\/strong>&nbsp;\u6e2c\u8a66\u4e2d\uff0c<strong>o1<\/strong>&nbsp;\u7372\u5f97\u4e86&nbsp;<strong>25%<\/strong>&nbsp;\u5230&nbsp;<strong>32%<\/strong>&nbsp;\u7684\u5206\u6578\uff08\u6eff\u5206&nbsp;<strong>100%<\/strong>\uff09\u3002\u96d6\u7136&nbsp;<strong>85%<\/strong>&nbsp;\u88ab\u8a8d\u70ba\u662f\u300c\u4eba\u985e\u6c34\u5e73\u300d\uff0c\u4f46&nbsp;<strong>ARC-AGI<\/strong>&nbsp;\u7684\u5275\u4f5c\u8005\u4e4b\u4e00&nbsp;<strong>Francois Chollet<\/strong>&nbsp;\u7a31\u9019\u4e00\u9032\u5c55\u70ba\u300c\u7a69\u5065\u300d\u3002\u7136\u800c\uff0c<strong>OpenAI<\/strong>&nbsp;\u8868\u793a\uff0c<strong>o3<\/strong>&nbsp;\u5728\u6700\u4f73\u60c5\u6cc1\u4e0b\u7372\u5f97\u4e86&nbsp;<strong>87.5%<\/strong>&nbsp;\u7684\u5206\u6578\uff0c\u5728\u6700\u5dee\u60c5\u6cc1\u4e0b\uff0c\u5176\u6027\u80fd\u662f&nbsp;<strong>o1<\/strong>&nbsp;\u7684\u4e09\u500d\u3002<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/arc-agi.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><figcaption class=\"wp-element-caption\"><strong>ARC-AGI<\/strong> \u6e2c\u8a66\u4e2d &#8211; <strong>85%<\/strong> \u88ab\u8a8d\u70ba\u662f\u9054\u5230\u4e86\u300c\u4eba\u985e\u6c34\u5e73\u300d<\/figcaption><\/figure>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/tentenai_-2024-12-21-at-9.46.19-AM.jpg?w=640&#038;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p>o3 \u7834\u7d00\u9304\u7684 ARC-AGI \u6027\u80fd\u65e2\u662f\u4e00\u500b\u91cc\u7a0b\u7891\uff0c\u4e5f\u662f\u4e00\u500b\u6311\u6230\uff0c\u70ba\u4eba\u5de5\u667a\u6167\u6240\u80fd\u5be6\u73fe\u7684\u76ee\u6a19\u8a2d\u5b9a\u4e86\u65b0\u7684\u6a19\u6e96\uff0c\u540c\u6642\u5f37\u8abf\u4e86\u5b83\u8ddd\u96e2&nbsp;\u901a\u7528\u4eba\u5de5\u667a\u6167 AGI&nbsp;\u9084\u6709\u591a\u9060\u3002<\/p>\n<\/blockquote>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>\u6a21\u578b\u540d\u7a31<\/th><th>\u516c\u958b\u8a55\u4f30\u5206\u6578<\/th><th>\u534a\u79c1\u4eba\u8a55\u4f30\u5206\u6578<\/th><th>\u5e73\u5747\u6bcf\u4efb\u52d9\u6642\u9593(\u5206\u9418)<\/th><\/tr><\/thead><tbody><tr><td>o3 (\u9ad8\u904b\u7b97)<\/td><td>&#8211;<\/td><td>87.5%<\/td><td>&#8211;<\/td><\/tr><tr><td>o3 (\u6a19\u6e96)<\/td><td>&#8211;<\/td><td>75.7%<\/td><td>&#8211;<\/td><\/tr><tr><td>o1-preview<\/td><td>21.2%<\/td><td>18%<\/td><td>4.2<\/td><\/tr><tr><td>Claude 3.5<\/td><td>21%<\/td><td>14%<\/td><td>0.3<\/td><\/tr><tr><td>o1-mini<\/td><td>12.8%<\/td><td>9.5%<\/td><td>3.0<\/td><\/tr><tr><td>GPT-4o<\/td><td>9%<\/td><td>5%<\/td><td>0.3<\/td><\/tr><tr><td>Gemini 1.5<\/td><td>8%<\/td><td>4.5%<\/td><td>1.1<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E9%87%8D%E8%A6%81%E7%AA%81%E7%A0%B4\">\u91cd\u8981\u7a81\u7834<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>o3 \u662f\u9996\u500b\u7a81\u7834 ARC-AGI \u57fa\u6e96\u6e2c\u8a66\u7684 AI \u6a21\u578b\uff0c\u6253\u7834\u4e86\u4e94\u5e74\u4f86\u7684\u7d00\u9304<\/li>\n\n\n\n<li>\u5728\u6a19\u6e96\u904b\u7b97\u6a21\u5f0f\u4e0b\u9054\u5230 75.7% \u7684\u5206\u6578\uff0c\u9ad8\u904b\u7b97\u6a21\u5f0f\u4e0b\u66f4\u9054\u5230 87.5%<\/li>\n\n\n\n<li>\u76f8\u6bd4\u4e4b\u4e0b\uff0cGPT-3 \u5728 2020 \u5e74\u7684\u5f97\u5206\u70ba 0%<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E6%AD%B7%E5%8F%B2%E9%80%B2%E5%B1%95\">\u6b77\u53f2\u9032\u5c55<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u5f9e 2020 \u5e74 GPT-3 \u7684 0% \u5230 2024 \u5e74 GPT-4o \u7684 5%\uff0c\u82b1\u4e86\u56db\u5e74\u6642\u9593<\/li>\n\n\n\n<li>2024 \u5e74: \u79c1\u4eba\u8a55\u4f30\u7684\u6700\u4f73\u8868\u73fe\u5f9e 33% \u63d0\u5347\u5230 55.5%<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E5%B0%88%E5%AE%B6%E8%A9%95%E5%83%B9\">\u5c08\u5bb6\u8a55\u50f9<\/h3>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Fran\u00e7ois Chollet \u6307\u51fa\uff0c\u901a\u904e ARC-AGI \u6e2c\u8a66\u4e26\u4e0d\u7b49\u540c\u65bc\u5be6\u73fe AGI<\/li>\n\n\n\n<li>\u5728\u5373\u5c07\u63a8\u51fa\u7684 ARC-AGI-2 \u57fa\u6e96\u6e2c\u8a66\u4e2d\uff0co3 \u7684\u8868\u73fe\u9810\u8a08\u6703\u964d\u81f3 30% \u4ee5\u4e0b\uff0c\u800c\u8070\u660e\u7684\u4eba\u985e\u4ecd\u53ef\u9054\u5230 95% \u4ee5\u4e0a\u7684\u5206\u6578<\/li>\n\n\n\n<li><strong>\u4e0b\u4e00\u4ee3\u57fa\u6e96\u6e2c\u8a66\u7684\u69cb\u5efa<\/strong><\/li>\n<\/ul>\n\n\n\n<p>\u503c\u5f97\u4e00\u63d0\u7684\u662f\uff0c<strong>OpenAI<\/strong>&nbsp;\u8868\u793a\u5c07\u8207&nbsp;<strong>ARC-AGI<\/strong>&nbsp;\u80cc\u5f8c\u7684\u57fa\u91d1\u6703\u5408\u4f5c\uff0c\u69cb\u5efa\u4e0b\u4e00\u4ee3\u57fa\u6e96\u6e2c\u8a66\uff0c\u9019\u5c07\u9032\u4e00\u6b65\u8a55\u4f30&nbsp;<strong>AI<\/strong>&nbsp;\u7cfb\u7d71\u5728\u7372\u53d6\u65b0\u6280\u80fd\u65b9\u9762\u7684\u80fd\u529b\u3002\u7576\u7136\uff0c<strong>ARC-AGI<\/strong>&nbsp;\u4e5f\u6709\u5176\u5c40\u9650\u6027\uff0c\u4e14\u5176\u5c0d&nbsp;<strong>AGI<\/strong>&nbsp;\u7684\u5b9a\u7fa9\u53ea\u662f\u773e\u591a\u5b9a\u7fa9\u4e2d\u7684\u4e00\u7a2e\u3002<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"o3-%E5%9C%A8%E5%85%B6%E4%BB%96%E5%9F%BA%E6%BA%96%E6%B8%AC%E8%A9%A6%E4%B8%AD%E7%9A%84%E8%A1%A8%E7%8F%BE\"><strong>o3 \u5728\u5176\u4ed6\u57fa\u6e96\u6e2c\u8a66\u4e2d\u7684\u8868\u73fe<\/strong><\/h3>\n\n\n\n<p>\u5728\u5176\u4ed6\u57fa\u6e96\u6e2c\u8a66\u4e2d\uff0c<strong>o3<\/strong>&nbsp;\u5c55\u73fe\u4e86\u5f37\u5927\u7684\u7af6\u722d\u529b\u3002\u5177\u9ad4\u8868\u73fe\u5982\u4e0b\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>SWE-Bench Verified<\/strong>\uff1a<strong>o3<\/strong>&nbsp;\u7684\u8868\u73fe\u512a\u65bc&nbsp;<strong>o1<\/strong>&nbsp;<strong>22.8 \u500b\u767e\u5206\u9ede<\/strong>\uff0c\u4e26\u7372\u5f97\u4e86&nbsp;<strong>2727 \u7684 Codeforces \u8a55\u7d1a<\/strong>\u3002<\/li>\n\n\n\n<li><strong>AIME 2024<\/strong>\uff1a<strong>o3<\/strong>&nbsp;\u7372\u5f97\u4e86&nbsp;<strong>96.7%<\/strong>&nbsp;\u7684\u5206\u6578\uff0c\u50c5\u932f\u4e86\u4e00\u500b\u554f\u984c\u3002<\/li>\n\n\n\n<li><strong>GPQA Diamond<\/strong>\uff1a<strong>o3<\/strong>&nbsp;\u7372\u5f97\u4e86&nbsp;<strong>87.7%<\/strong>&nbsp;\u7684\u5206\u6578\u3002<\/li>\n\n\n\n<li><strong>EpochAI \u7684 Frontier Math<\/strong>\uff1a<strong>o3<\/strong>&nbsp;\u89e3\u6c7a\u4e86&nbsp;<strong>25.2%<\/strong>&nbsp;\u7684\u554f\u984c\uff0c\u7121\u5176\u4ed6\u6a21\u578b\u8d85\u904e&nbsp;<strong>2%<\/strong>\u3002<\/li>\n<\/ul>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/GfQrwytXUAENEfk.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/GfQsAg_WUAERy47.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/GfQsIkEWAAAby-8.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/GfQsTFvXwAAjaq_.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><\/figure>\n<\/div>\n\n\n<p>\u9019\u4e9b\u6578\u64da\u986f\u793a\uff0c<strong>o3<\/strong>&nbsp;\u5728\u5df2\u77e5\u7684\u6700\u56f0\u96e3\u8a55\u4f30\u4e2d\u5275\u4e0b\u4e86\u65b0\u7d00\u9304\uff0c\u5c55\u793a\u4e86\u5176\u5353\u8d8a\u7684\u63a8\u7406\u548c\u89e3\u6c7a\u554f\u984c\u7684\u80fd\u529b\u3002\u7136\u800c\uff0c\u9019\u4e9b\u7d50\u679c\u4f86\u81ea&nbsp;<strong>OpenAI<\/strong>&nbsp;\u7684\u5167\u90e8\u8a55\u4f30\uff0c\u5c1a\u9700\u7b49\u5f85\u5916\u90e8\u5ba2\u6236\u548c\u7d44\u7e54\u7684\u57fa\u6e96\u6e2c\u8a66\u4f86\u9032\u4e00\u6b65\u9a57\u8b49\u3002<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter\"><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/s4.tenten.co\/learning\/content\/images\/2024\/12\/kwdo648cx18e1.jpeg?w=640&#038;ssl=1\" alt=\"\"\/><figcaption class=\"wp-element-caption\">OpenAI \u7684 o3 \u6a21\u578b\u5728\u7af6\u6280\u7a0b\u5f0f\u8a2d\u8a08\u9818\u57df\u5df2\u9054\u5230\u5168\u7403\u6392\u540d\u7b2c 175 \u540d\u7684\u4eba\u985e\u9078\u624b\u6c34\u6e96\uff01<\/figcaption><\/figure>\n<\/div>\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E6%8E%A8%E7%90%86%E6%A8%A1%E5%9E%8B%E7%9A%84%E6%9C%AA%E4%BE%86%E8%B6%A8%E5%8B%A2\"><strong>\u63a8\u7406\u6a21\u578b\u7684\u672a\u4f86\u8da8\u52e2<\/strong><\/h3>\n\n\n\n<p>\u81ea&nbsp;<strong>OpenAI<\/strong>&nbsp;\u767c\u5e03\u5176\u9996\u7cfb\u5217\u63a8\u7406\u6a21\u578b\u4ee5\u4f86\uff0c\u7af6\u722d\u5c0d\u624b\u7684&nbsp;<strong>AI<\/strong>&nbsp;\u516c\u53f8\u7d1b\u7d1b\u63a8\u51fa\u4e86\u5927\u91cf\u7684\u63a8\u7406\u6a21\u578b\uff0c\u5305\u62ec&nbsp;<strong>Google<\/strong>\u3002<strong>11 \u6708\u521d<\/strong>\uff0c\u7531\u91cf\u5316\u4ea4\u6613\u54e1\u8cc7\u52a9\u7684&nbsp;<strong>AI<\/strong>&nbsp;\u7814\u7a76\u516c\u53f8&nbsp;<strong>DeepSeek<\/strong>&nbsp;\u767c\u5e03\u4e86\u5176\u9996\u500b\u63a8\u7406\u6a21\u578b&nbsp;<strong>DeepSeek-R1<\/strong>&nbsp;\u7684\u9810\u89bd\u7248\u3002\u540c\u6708\uff0c<strong>\u963f\u91cc\u5df4\u5df4<\/strong>&nbsp;\u7684&nbsp;<strong>Qwen \u5718\u968a<\/strong>&nbsp;\u516c\u5e03\u4e86\u64da\u7a31\u662f&nbsp;<strong>o1<\/strong>&nbsp;\u7684\u9996\u500b\u300c\u958b\u653e\u300d\u6311\u6230\u8005\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"%E6%8E%A8%E7%90%86%E6%A8%A1%E5%9E%8B%E7%9A%84%E9%96%8B%E7%99%BC%E5%8B%95%E5%8A%9B\"><strong>\u63a8\u7406\u6a21\u578b\u7684\u958b\u767c\u52d5\u529b<\/strong><\/h4>\n\n\n\n<p>\u63a8\u7406\u6a21\u578b\u7684\u8208\u8d77\uff0c\u4e3b\u8981\u6e90\u65bc\u5c0b\u627e\u6539\u9032\u751f\u6210\u5f0f&nbsp;<strong>AI<\/strong>&nbsp;\u7684\u65b0\u65b9\u6cd5\u3002\u9019\u4e9b\u6a21\u578b\u80fd\u5920\u66f4\u6709\u6548\u5730\u8655\u7406\u8907\u96dc\u7684\u554f\u984c\uff0c\u63d0\u4f9b\u66f4\u7cbe\u78ba\u7684\u89e3\u7b54\u3002\u7136\u800c\uff0c\u4e26\u975e\u6240\u6709\u4eba\u90fd\u76f8\u4fe1\u63a8\u7406\u6a21\u578b\u662f\u524d\u9032\u7684\u6700\u4f73\u9053\u8def\u3002\u4e00\u65b9\u9762\uff0c\u904b\u884c\u9019\u4e9b\u6a21\u578b\u9700\u8981\u5927\u91cf\u7684\u8a08\u7b97\u80fd\u529b\uff0c\u5c0e\u81f4\u5176\u6210\u672c\u8f03\u9ad8\uff1b\u53e6\u4e00\u65b9\u9762\uff0c\u96d6\u7136\u76ee\u524d\u5b83\u5011\u5728\u57fa\u6e96\u6e2c\u8a66\u4e2d\u8868\u73fe\u51fa\u8272\uff0c\u4f46\u5c1a\u4e0d\u6e05\u695a\u63a8\u7406\u6a21\u578b\u80fd\u5426\u6301\u7e8c\u4fdd\u6301\u9019\u7a2e\u9032\u5c55\u901f\u5ea6\u3002<\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"o3-%E7%99%BC%E5%B8%83%E6%99%82%E7%9A%84%E5%85%B6%E4%BB%96%E6%96%B0%E8%81%9E\"><strong>o3 \u767c\u5e03\u6642\u7684\u5176\u4ed6\u65b0\u805e<\/strong><\/h4>\n\n\n\n<p>\u6709\u8da3\u7684\u662f\uff0c<strong>o3<\/strong>&nbsp;\u7684\u767c\u5e03\u6070\u9022&nbsp;<strong>OpenAI<\/strong>&nbsp;\u6700\u6709\u6210\u5c31\u7684\u79d1\u5b78\u5bb6\u4e4b\u4e00&nbsp;<strong>Alec Radford<\/strong>&nbsp;\u96e2\u8077\u4e4b\u969b\u3002<strong>Radford<\/strong>&nbsp;\u662f&nbsp;<strong>OpenAI\u300cGPT \u7cfb\u5217\u300d\u751f\u6210\u5f0f AI \u6a21\u578b\uff08\u5982 GPT-3\u3001GPT-4 \u7b49\uff09<\/strong>&nbsp;\u7684\u4e3b\u8981\u4f5c\u8005\uff0c\u672c\u9031\u5ba3\u5e03\u5c07\u96e2\u958b&nbsp;<strong>OpenAI<\/strong>\uff0c\u8f49\u800c\u9032\u884c\u7368\u7acb\u7814\u7a76\u3002\u9019\u4e00\u8b8a\u52d5\u7121\u7591\u70ba&nbsp;<strong>OpenAI<\/strong>&nbsp;\u7684\u672a\u4f86\u767c\u5c55\u589e\u6dfb\u4e86\u65b0\u7684\u8b8a\u6578\u3002<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"%E7%B5%90%E8%AA%9E\"><strong>\u7d50\u8a9e<\/strong><\/h3>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"o3-%E7%99%BC%E5%B8%83%E6%99%82%E9%96%93\">o3 \u767c\u5e03\u6642\u9593<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>o3-mini \u9810\u8a08\u5c07\u65bc 2024 \u5e74 1 \u6708\u5e95\u63a8\u51fa<\/li>\n\n\n\n<li>\u5b8c\u6574\u7248 o3 \u7684\u5177\u9ad4\u767c\u5e03\u65e5\u671f\u5c1a\u672a\u516c\u5e03\uff0c\u4f46\u6703\u5728 o3-mini \u4e4b\u5f8c\u63a8\u51fa<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E6%88%90%E6%9C%AC%E8%B3%87%E8%A8%8A\">\u6210\u672c\u8cc7\u8a0a<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u4f4e\u904b\u7b97\u6a21\u5f0f\u4e0b\uff0c\u6bcf\u500b\u4efb\u52d9\u7684\u6210\u672c\u7d04\u70ba $17-20<\/li>\n\n\n\n<li>\u9ad8\u904b\u7b97\u6a21\u5f0f\uff08\u6bd4\u6a19\u6e96\u7248\u672c\u9ad8 172 \u500d\u7684\u904b\u7b97\u80fd\u529b\uff09\u7684\u6210\u672c\u5c1a\u672a\u516c\u958b<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E4%B8%BB%E8%A6%81%E6%80%A7%E8%83%BD%E6%8F%90%E5%8D%87\">\u4e3b\u8981\u6027\u80fd\u63d0\u5347<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u5728\u5e38\u898b\u7a0b\u5f0f\u8a2d\u8a08\u4efb\u52d9\u4e2d\uff0c\u6e96\u78ba\u7387\u6bd4 o1 \u63d0\u5347\u8d85\u904e 20%<\/li>\n\n\n\n<li>\u5728 ARC-AGI \u8a55\u4f30\u4e2d\uff0c\u4f4e\u904b\u7b97\u7248\u672c\u9054\u5230 75.7%\uff0c\u9ad8\u904b\u7b97\u7248\u672c\u9054\u5230 87.5% \u7684\u5206\u6578<\/li>\n\n\n\n<li>\u5728 AIME 2024 \u6578\u5b78\u6e2c\u9a57\u4e2d\uff0c\u6e96\u78ba\u7387\u9054\u5230 96.7%\uff0c\u76f8\u6bd4 o1 \u7684 83.3%<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E6%96%B0%E5%8A%9F%E8%83%BD%E7%89%B9%E9%BB%9E\">\u65b0\u529f\u80fd\u7279\u9ede<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u53ef\u8abf\u6574\u63a8\u7406\u6642\u9593<\/strong>\uff1a\u63d0\u4f9b\u4f4e\u3001\u4e2d\u3001\u9ad8\u4e09\u7a2e\u904b\u7b97\u6a21\u5f0f\uff0c\u4f7f\u7528\u8005\u53ef\u6839\u64da\u9700\u6c42\u8abf\u6574\u601d\u8003\u6642\u9593<\/li>\n\n\n\n<li><strong>\u7a0b\u5f0f\u641c\u5c0b\u80fd\u529b<\/strong>\uff1a\u63a1\u7528\u6df1\u5ea6\u5b78\u7fd2\u5f15\u5c0e\u7684\u7a0b\u5f0f\u641c\u5c0b\u65b9\u5f0f\uff0c\u80fd\u5728\u57f7\u884c\u6642\u91cd\u7d44\u77e5\u8b58<\/li>\n\n\n\n<li><strong>\u9069\u61c9\u6027\u601d\u7dad<\/strong>\uff1a\u80fd\u5920\u8655\u7406\u524d\u6240\u672a\u898b\u7684\u4efb\u52d9\uff0c\u63a5\u8fd1\u4eba\u985e\u6c34\u5e73\u7684\u8868\u73fe<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E6%A8%A1%E5%9E%8B%E8%AE%8A%E9%AB%94\">\u6a21\u578b\u8b8a\u9ad4<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>o3-mini<\/strong>\uff1a\n<ul class=\"wp-block-list\">\n<li>\u6027\u80fd\u7565\u512a\u65bc o1<\/li>\n\n\n\n<li>\u5ef6\u9072\u548c\u56de\u61c9\u6642\u9593\u8207\u6a19\u6e96\u6a21\u578b\u76f8\u7576<\/li>\n\n\n\n<li>\u9810\u8a08\u65bc 2024 \u5e74 1 \u6708\u63a8\u51fa<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E5%AE%89%E5%85%A8%E6%80%A7%E6%94%B9%E9%80%B2\">\u5b89\u5168\u6027\u6539\u9032<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u63a1\u7528\u6df1\u601d\u719f\u616e\u7684\u5c0d\u9f4a\u8a13\u7df4\u65b9\u5f0f<\/li>\n\n\n\n<li>\u5728\u8655\u7406\u60e1\u610f\u63d0\u793a\u548c\u826f\u6027\u63d0\u793a\u65b9\u9762\u90fd\u6709\u6240\u6539\u9032<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E7%9B%AE%E5%89%8D%E7%8B%80%E6%85%8B\">\u76ee\u524d\u72c0\u614b<\/h5>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\u6a21\u578b\u73fe\u6b63\u9032\u884c\u516c\u5171\u5b89\u5168\u8a55\u4f30\u968e\u6bb5<\/li>\n\n\n\n<li>\u5b89\u5168\u548c\u5b89\u5168\u7814\u7a76\u4eba\u54e1\u53ef\u4ee5\u8a3b\u518a\u7533\u8acb\u9810\u89bd\u548c\u8a55\u4f30\u9019\u4e9b\u6a21\u578b<\/li>\n<\/ul>\n\n\n\n<h5 class=\"wp-block-heading\" id=\"%E6%B8%AC%E8%A9%A6%E8%A1%A8%E7%8F%BE\">\u6e2c\u8a66\u8868\u73fe<\/h5>\n\n\n\n<p>\u2022 ARC-AGI \u6e2c\u8a66\uff1ao3 \u4ee5\u4f4e\u904b\u7b97\u8cc7\u6e90\u9054\u6210\u8d85\u8d8a o1 \u4e09\u500d\u4ee5\u4e0a\u7684\u5206\u6578\uff0c\u7e3d\u5206\u7a81\u7834 87%<\/p>\n\n\n\n<p>\u2022 EpochAI \u524d\u6cbf\u6578\u5b78\uff1a\u5275\u4e0b 25.2% \u7684\u89e3\u984c\u7d00\u9304\uff0c\u800c\u5176\u4ed6\u6a21\u578b\u5747\u672a\u8d85\u904e 2%<\/p>\n\n\n\n<p>\u2022 SWE-Bench \u7a0b\u5f0f\u9a57\u8b49\uff1a\u6bd4 o1 \u63d0\u5347\u4e86 22.8 \u500b\u767e\u5206\u9ede<\/p>\n\n\n\n<p>\u2022 Codeforces \u7af6\u8cfd\uff1a\u9054\u5230 2727 \u5206\uff0c\u8d85\u8d8a\u4e86 OpenAI \u9996\u5e2d\u79d1\u5b78\u5bb6\u7684 2665 \u5206<\/p>\n\n\n\n<p>\u2022 AIME 2024 \u6578\u5b78\u7af6\u8cfd\uff1a\u9a5a\u4eba\u7684 96.7% \u6b63\u78ba\u7387\uff0c\u50c5\u932f\u4e00\u984c<\/p>\n\n\n\n<p>\u2022 GPQA Diamond \u6e2c\u8a66\uff1a\u9054\u6210 87.7% \u7684\u6210\u7e3e\uff0c\u9060\u8d85\u4eba\u985e\u5c08\u5bb6\u6c34\u5e73<\/p>\n\n\n\n<p><strong>OpenAI<\/strong>&nbsp;\u63a8\u51fa\u7684&nbsp;<strong>o3 \u6a21\u578b<\/strong>&nbsp;\u7121\u7591\u662f\u4eba\u5de5\u667a\u80fd\u9818\u57df\u7684\u4e00\u5927\u7a81\u7834\u3002\u5176\u5f37\u5927\u7684\u63a8\u7406\u80fd\u529b\u3001\u9748\u6d3b\u7684\u61c9\u7528\u9078\u9805\u4ee5\u53ca\u5728\u591a\u9805\u57fa\u6e96\u6e2c\u8a66\u4e2d\u7684\u512a\u7570\u8868\u73fe\uff0c\u5c55\u793a&nbsp;<strong>OpenAI<\/strong>&nbsp;\u5728\u8ffd\u6c42\u66f4\u9ad8\u5c64\u6b21\u667a\u80fd\u65b9\u9762\u7684\u6c7a\u5fc3\u8207\u5be6\u529b\u3002<\/p>\n\n\n\n<p>\u8cc7\u6599\u4f86\u6e90: <a href=\"https:\/\/tenten.co\/learning\/openai-o3\/\">https:\/\/tenten.co\/learning\/openai-o3\/<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>2024-12-21 | Erik OpenAI \u63a8\u51fa\u8fc4\u4eca\u70ba\u6b62\u6700\u5f37\u5927\u7684 AI \u6a21\u578b\u7684\u6539\u9032\u7248\u672c o3 OpenAI&nbsp;\u5728\u5176\u70ba\u671f&nbsp;12 \u5929&nbsp;\u7684\u300cshipma&hellip;<\/p>\n","protected":false},"author":4,"featured_media":7903,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[579],"tags":[26],"class_list":["post-7902","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-579","tag-ai"],"gutentor_comment":0,"jetpack_featured_media_url":"https:\/\/i0.wp.com\/aict.nkust.edu.tw\/digitrans\/wp-content\/uploads\/2024\/12\/%E8%9E%A2%E5%B9%95%E6%93%B7%E5%8F%96%E7%95%AB%E9%9D%A2-2024-12-27-150212.png?fit=1059%2C587&ssl=1","jetpack-related-posts":[],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/posts\/7902","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=7902"}],"version-history":[{"count":3,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/posts\/7902\/revisions"}],"predecessor-version":[{"id":7935,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/posts\/7902\/revisions\/7935"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=\/wp\/v2\/media\/7903"}],"wp:attachment":[{"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=7902"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=7902"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aict.nkust.edu.tw\/digitrans\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=7902"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}