{"id":49753,"date":"2026-02-06T11:50:12","date_gmt":"2026-02-06T03:50:12","guid":{"rendered":"https:\/\/www.1ai.net\/?p=49753"},"modified":"2026-02-06T11:50:12","modified_gmt":"2026-02-06T03:50:12","slug":"openai-%e6%9c%80%e5%bc%ba-ai-%e6%99%ba%e8%83%bd%e4%bd%93%e7%bc%96%e7%a8%8b%e6%a8%a1%e5%9e%8b%ef%bc%9agpt-5-3-codex-%e7%99%bb%e5%9c%ba%ef%bc%8c%e5%9f%ba%e5%87%86%e6%b5%8b%e8%af%95%e5%85%a8%e9%9d%a2","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/49753.html","title":{"rendered":"OpenAI Best AI Smart Body Programming Model: GPT-5.3-Codex presentation, full-scale benchmarking"},"content":{"rendered":"<p>2 \u6708 6 \u65e5\u6d88\u606f\uff0c<a href=\"https:\/\/www.1ai.net\/en\/tag\/openai\" title=\"\u3010\u67e5\u770b\u542b\u6709[OpenAI]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">OpenAI<\/a> \u6628\u65e5\uff082 \u6708 5 \u65e5\uff09\u53d1\u5e03\u516c\u544a\uff0c<strong>\u5ba3\u5e03\u63a8\u51fa GPT-5.3-Codex\uff0c\u5ba3\u79f0\u8fd9\u662f\u5176\u8fc4\u4eca\u6700\u5f3a\u7684 AI <a href=\"https:\/\/www.1ai.net\/en\/tag\/%e6%99%ba%e8%83%bd%e4%bd%93\" title=\"\u3010\u67e5\u770b\u542b\u6709[\u667a\u80fd\u4f53]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">\u667a\u80fd\u4f53<\/a><a href=\"https:\/\/www.1ai.net\/en\/tag\/%e7%bc%96%e7%a8%8b%e6%a8%a1%e5%9e%8b\" title=\"\u3010\u67e5\u770b\u542b\u6709[\u7f16\u7a0b\u6a21\u578b]\u6807\u7b7e\u7684\u6587\u7ae0\u3011\" target=\"_blank\" rel=\"noopener\">\u7f16\u7a0b\u6a21\u578b<\/a>\u3002<\/strong><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-49754\" title=\"f1bc56f3j00ta0qn1001td000uy00t8p\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2026\/02\/f1bc56f3j00ta0qn1001td000uy00t8p.jpg\" alt=\"f1bc56f3j00ta0qn1001td000uy00t8p\" width=\"1114\" height=\"1052\" \/><\/p>\n<p>1AI\u9644\u4e0a\u76f8\u5173\u57fa\u51c6\u6d4b\u8bd5\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<ul>\n<li>\u5728 SWE-bench Pro\uff08\u516c\u5f00\u7248\uff09\u6d4b\u8bd5\u4e2d\uff0c\u8be5\u6a21\u578b\u5f97\u5206 56.8%\uff0c\u8d85\u8d8a\u4e86\u524d\u4ee3\u7684 56.4%\uff1b<\/li>\n<li>\u5728 Terminal-Bench 2.0 \u6d4b\u8bd5\u4e2d\uff0c\u5176\u5f97\u5206\u4ece\u524d\u4ee3\u7684 64.0% \u5927\u5e45\u8dc3\u5347\u81f3 77.3%\uff1b<\/li>\n<li>\u6700\u4e3a\u663e\u8457\u7684\u662f\u5728 OSWorld-Verified \u667a\u80fd\u4f53\u8ba1\u7b97\u673a\u64cd\u4f5c\u57fa\u51c6\u6d4b\u8bd5\u4e2d\uff0c\u5f97\u5206\u4ece 38.2% \u98d9\u5347\u81f3 64.7%\uff0c\u8868\u660e\u5176\u5728\u590d\u6742\u73af\u5883\u4e0b\u7684\u64cd\u4f5c\u80fd\u529b\u5b9e\u73b0\u4e86\u8d28\u7684\u98de\u8dc3\u3002<\/li>\n<li>\u5b98\u65b9\u6570\u636e\u663e\u793a\uff0c\u8be5\u6a21\u578b\u4e0d\u4ec5\u5728\u591a\u9879\u7f16\u7a0b\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u521b\u4e0b\u65b0\u9ad8\uff0c\u8fd0\u884c\u901f\u5ea6\u4e5f\u6bd4\u524d\u4ee3 GPT-5.2-Codex \u63d0\u5347\u4e86 25%\u3002<\/li>\n<\/ul>\n<p>OpenAI \u81f4\u529b\u4e8e\u5c06 GPT-5.3-Codex \u6253\u9020\u4e3a\u66f4\u4f18\u79c0\u7684\u5f00\u53d1\u8005\u534f\u4f5c\u5de5\u5177\u3002\u65b0\u6a21\u578b\u652f\u6301\u7528\u6237\u5728\u4efb\u52a1\u6267\u884c\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u201c\u5f15\u5bfc\u4e0e\u4ea4\u4e92\u201d\uff0c\u4e14\u4e0d\u4f1a\u5bfc\u81f4\u4e0a\u4e0b\u6587\u4e22\u5931\u3002<\/p>\n<p>\u5728 Codex \u5e94\u7528\u4e2d\uff0c\u6a21\u578b\u4f1a\u9ad8\u9891\u66f4\u65b0\u5de5\u4f5c\u8fdb\u5ea6\uff0c\u5141\u8bb8\u5f00\u53d1\u8005\u5b9e\u65f6\u63d0\u95ee\u3001\u8ba8\u8bba\u65b9\u6848\u5e76\u4fee\u6b63\u65b9\u5411\u3002\u5b98\u65b9\u535a\u6587\u6307\u51fa\uff0cCodex \u6b63\u4ece\u5355\u4e00\u7684\u7f16\u7a0b\u667a\u80fd\u4f53\u6f14\u53d8\u4e3a\u66f4\u901a\u7528\u7684\u8ba1\u7b97\u673a\u534f\u4f5c\u8005\uff0c\u5927\u5e45\u964d\u4f4e\u4e86\u6784\u5efa\u5e94\u7528\u7684\u95e8\u69db\u3002<\/p>\n<p>GPT-5.3-Codex \u76ee\u524d\u5df2\u5168\u9762\u5411 ChatGPT \u4ed8\u8d39\u7528\u6237\u5f00\u653e\uff0c\u6db5\u76d6 Plus\u3001Team \u53ca Enterprise \u7b49\u8ba2\u9605\u8ba1\u5212\u3002\u7528\u6237\u53ef\u901a\u8fc7 Codex \u5e94\u7528\u3001\u547d\u4ee4\u884c\u5de5\u5177\uff08CLI\uff09\u3001IDE \u6269\u5c55\u63d2\u4ef6\u53ca\u7f51\u9875\u7aef\u7acb\u5373\u8bbf\u95ee\uff0cAPI \u7248\u672c\u4e5f\u5c06\u5728\u8fd1\u671f\u4e0a\u7ebf\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>On February 6, OpenAI announced yesterday (5 February) the launch of GPT-5.3-Codex, claiming it was its strongest AI intelligent body programming model to date. (a) In the SWE-bench Pro (open version) test, the model scored 56.8%, exceeding the previous generation's 56.4%<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[146],"tags":[190,1355,7692],"collection":[],"class_list":["post-49753","post","type-post","status-publish","format-standard","hentry","category-news","tag-openai","tag-1355","tag-7692"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/49753","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=49753"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/49753\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=49753"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=49753"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=49753"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=49753"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}