{"id":47287,"date":"2025-12-12T11:56:01","date_gmt":"2025-12-12T03:56:01","guid":{"rendered":"https:\/\/www.1ai.net\/?p=47287"},"modified":"2025-12-12T11:56:01","modified_gmt":"2025-12-12T03:56:01","slug":"openai-%e6%8e%a8%e5%87%ba-gpt-5-2-%e6%ad%a3%e5%bc%8f%e5%8f%91%e5%b8%83%ef%bc%8c%e7%8b%99%e5%87%bb-google-gemini-3","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/47287.html","title":{"rendered":"OpenAI Launch GPT-5.2 Official launch, sniper Google Gemini 3"},"content":{"rendered":"

December 12th news, this morningOpenAI<\/a> OFFICIAL PUBLICATION OF THE GPT-5.2 SERIES OF MODELS FOR ChatGPT<\/a> Fee-paying users and developers API, in three versions: Instant, Thinking, Pro, located in a more reliable production-level model\u3002<\/p>\n

\"1ce66d50j00t751iz005d000cs007bp\"<\/p>\n

Instant Edition: Emphasizing low delay and high response, adapting to routine tasks such as information queries, document translation, basic writing, etc<\/p>\n

Thinking Edition: Focus programming, long document analysis, mathematical reasoning and project planning, smart assistant to locate complex enterprise-level scenarios<\/p>\n

Pro Edition: For scientific research, finance and difficult tasks, with emphasis on extreme accuracy and reliability, called the \"Capacity ceiling\"<\/p>\n

IN TERMS OF TECHNICAL PERFORMANCE, GPT-5.2 HAS BEEN SIGNIFICANTLY ENHANCED IN PROGRAMMING, MATHEMATICAL AND SCIENTIFIC COMPUTING, FACTUAL ACCURACY AND MULTI-MODULAR PROCESSING, ENABLING THE DIRECT GENERATION OF TABLES, THE PRODUCTION OF PPTS AND THE DEVELOPMENT OF MULTILINGUAL CODES, FURTHER ENHANCING ITS PRODUCTIVITY POSITION IN PROFESSIONAL SETTINGS\u3002<\/p>\n

At the same time, OpenAI announced plans to go online in the first quarter of 2026 and update the knowledge base to August 2025 to ensure timeliness and accuracy of information\u3002<\/p>\n

Before, in the face of Google Gemini 3, OpenAI CEO Otman had opened the Red Alert inside the company\u3002<\/p>\n

He acknowledged that Google ' s success in pre-training had narrowed the gap between the two sides and could bring economic pressure. To this end, OpenAI is adjusting its strategy to temporarily slow down the AGI target and to do its best to improve the quality of ChatGPT to cope with competition\u3002<\/p>\n

In an interview with CNBC on Thursday, Otman local time stated that Google\u2019s \u201cGemini 3\u201d impact on corporate indicators was lower than expected and that OpenAI was expected to withdraw from the \u201cRed Alert\u201d in January\u3002<\/p>\n

The OpenAI application CEO Fidji Simo states that the release was not driven by competition this week, but rather by the results of months of consolidation\u3002<\/p>\n

GPT-5.2 is now in the ChatGPT and gives priority to paying users; GPT-51 will remain in the \"Traditional Model\" option for three months to go down. API is open simultaneously at higher prices than GPT-51, but the total cost is expected to be lower because of increased token efficiency\u3002<\/p>\n

Today, OpenAI has a three-year mandate agreement with Disney. Users can generate social videos with more than 200 characters, including Disney, Manover, Pixar and Star Wars, and some videos can be shown on Disney+\u3002<\/p>\n

In exchange, Disney invested $1 billion in OpenAI and will be an important client of the latter\u3002<\/p>","protected":false},"excerpt":{"rendered":"

On December 12th, in the morning, OpenAI officially released the GPT-5.2 series of models for ChatGPT paid users and developers API, distributed in Instant, Thinking and Pro versions, which are positioned as more reliable production-grade models. Instant Edition: Emphasizing Low Delays and High Response, Fitting to Regular Tasks such as Information Query, Document Translation, Basic Writing, Thinking Edition: Focusing Programming, Long Document Analysis, Mathematic Logic and Project Planning, Smart Assistant for Positioning complex enterprise-level scenes Pro Edition: Focusing on Research, Finance and Difficulty Tasks, Emphasizing Extreme Accuracy and Reliability, known as the \"Capacity ceiling\" Technically, GPT-5.2 is programmed<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[177,190],"collection":[],"class_list":["post-47287","post","type-post","status-publish","format-standard","hentry","category-news","tag-chatgpt","tag-openai"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/47287","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=47287"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/47287\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=47287"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=47287"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=47287"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=47287"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}