{"id":46813,"date":"2025-12-03T11:42:28","date_gmt":"2025-12-03T03:42:28","guid":{"rendered":"https:\/\/www.1ai.net\/?p=46813"},"modified":"2025-12-03T11:42:28","modified_gmt":"2025-12-03T03:42:28","slug":"openai-%e8%a2%ab%e6%9b%9d%e6%ad%a3%e5%bc%80%e5%8f%91%e5%85%a8%e6%96%b0-ai%e5%a4%a7%e8%af%ad%e8%a8%80%e6%a8%a1%e5%9e%8b%ef%bc%9a%e4%bb%a3%e5%8f%b7%e5%a4%a7%e8%92%9c%ef%bc%8c%e6%80%a7","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/46813.html","title":{"rendered":"OpenAI is being exposed to the development of a new AI large-language model: codenamed Garlic, performance beyond Google Gemini 3"},"content":{"rendered":"<p>On December 3rd, the Information reported that<a href=\"https:\/\/www.1ai.net\/en\/tag\/openai\" title=\"[View articles tagged with [OpenAI]]\" target=\"_blank\" >OpenAI<\/a> NEW AI UNDER DEVELOPMENT <a href=\"https:\/\/www.1ai.net\/en\/tag\/%e5%a4%a7%e8%af%ad%e8%a8%80%e6%a8%a1%e5%9e%8b\" title=\"[View articles tagged with [large language model]]\" target=\"_blank\" >Large Language Model<\/a>In response to the technical competition of Google Gemini 3, the project is codenamed \"Greek.\"<a href=\"https:\/\/www.1ai.net\/en\/tag\/garlic\" title=\"[See articles with [Garlic] labels]\" target=\"_blank\" >Garlic<\/a>).<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-46814\" title=\"423953d8j00t6ocxg0019d000g600j2p\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2025\/12\/423953d8j00t6ocxg0019d000g600j2p.jpg\" alt=\"423953d8j00t6ocxg0019d000g600j2p\" width=\"582\" height=\"686\" \/><\/p>\n<p>According to the report, the Chief Research Officer of OpenAI, Mark Chen, introduced the new one to the team last week <a href=\"https:\/\/www.1ai.net\/en\/tag\/ai%e6%a8%a1%e5%9e%8b\" title=\"[View articles tagged with [AI models]]\" target=\"_blank\" >AI Models<\/a>It is alleged that the model performed better than Google Gemini 3.0 and Anthropic Opus 4.5 in its programming and logical reasoning tasks\u3002<\/p>\n<p>The CEO of OpenAI, Sam Altmann, had previously launched an emergency response mechanism to enhance ChatGPT performance and planned to take the lead in launching a new model that optimized reasoning\u3002<\/p>\n<p>According to the Information, \u201cGarlic\u201d is a different technical route from the previously disclosed \u201cShallotpeat\u201d and its core breakthrough lies in the optimization of the pre-training phase: a significant reduction in training costs and time by improving the algorithm architecture to achieve the injection of previously large-scale model-based knowledge density into smaller-parametric models. The technical path directly addressed the \u201cpre-training leap\u201d claimed by Google and repaired the key deficiencies of GPT-4.5. OpenAI indicates that, based on this experience, the development of larger models has been initiated\u3002<\/p>\n<p>The current \u201cGarlic\u201d still needs to complete a follow-up process such as security assessment, fine-tuning in specific areas (notes: medical, legal), which is expected to be published in GPT-5.2 or GPT-5.5 by early 2026\u3002<\/p>","protected":false},"excerpt":{"rendered":"<p>On December 3rd, The Information reported that OpenAI was developing a new AI Large Language Model to respond to the technical competition of Google Gemini 3, codenamed Garlic. It was described that the new AI model, presented to the team last week by Mark Chen, Chief Research Officer of OpenAI, was said to have performed better than Google Gemini 3.0 and Anthropic Opus 4.5 in the programming and logical reasoning mission. OpenAI Chief Executive Officer Sam Altmann has previously launched an emergency response mechanism to enhance ChatGPT performance and plans to take the lead in launching a new model for optimal reasoning<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[167,7946,190,706],"collection":[],"class_list":["post-46813","post","type-post","status-publish","format-standard","hentry","category-news","tag-ai","tag-garlic","tag-openai","tag-706"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/46813","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=46813"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/46813\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=46813"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=46813"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=46813"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=46813"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}