{"id":33457,"date":"2025-11-13T09:30:34","date_gmt":"2025-11-13T09:30:34","guid":{"rendered":"https:\/\/metaverseplanet.net\/blog\/?p=33457"},"modified":"2026-01-05T08:01:14","modified_gmt":"2026-01-05T08:01:14","slug":"openai-introduces-gpt-5-1","status":"publish","type":"post","link":"https:\/\/metaverseplanet.net\/blog\/openai-introduces-gpt-5-1\/","title":{"rendered":"OpenAI introduces GPT 5.1: Smarter and more conversation-focused"},"content":{"rendered":"\n

OpenAI<\/strong> has announced GPT-5.1<\/strong>, an updated version of its flagship model, GPT-5<\/strong>. The company states the new version is smarter and more conversation-focused.<\/p>\n\n\n\n

OpenAI today announced GPT-5.1<\/strong>, an updated version of its flagship model GPT-5<\/strong>, which was released in August. The company describes this version as an upgrade that makes ChatGPT<\/strong> smarter and more enjoyable to chat with.<\/p>\n\n\n\n

\n\n\n\n

What’s new in GPT-5.1?<\/h2>\n\n\n\n
\"\"<\/figure>\n\n\n\n

The new version comes with two models: GPT-5.1 Instant<\/strong> and GPT-5.1 Thinking<\/strong>. According to OpenAI, the Instant model<\/strong> is the first to use adaptive reasoning<\/strong> to decide when it needs to “think” before answering more challenging questions, thereby providing more comprehensive and accurate answers while also maintaining speed.<\/p>\n\n\n\n

The Thinking model<\/strong>, on the other hand, has been made more stable on complex tasks, faster on simple tasks, and more understandable. The model now adapts its “thinking time” more precisely to the question, dedicating more time to complex problems while responding faster to simpler ones.<\/p>\n\n\n\n

OpenAI is also expanding the model’s conversational style options<\/strong> with the new update. Options now include tones such as Default<\/strong>, Professional<\/strong>, Friendly<\/strong>, Candid<\/strong>, Quirky<\/strong>, Efficient<\/strong>, Nerdy<\/strong>, and Cynical<\/strong>. The company also announced that a new experimental feature<\/strong> allowing users to customize ChatGPT’s style<\/strong> directly from the settings will roll out to some users this week.<\/p>\n\n\n\n

\n\n\n\n

The new models will begin rolling out this week. The old GPT-5<\/strong> models will remain in ChatGPT\u2019s \u201clegacy models<\/strong>\u201d menu for three months before being removed.<\/p>\n\n\n\n

GPT-5 failed to meet expectations<\/h2>\n\n\n\n
\"\"<\/figure>\n\n\n\n

Although GPT-5<\/strong> was introduced in August with high expectations, many ChatGPT<\/a><\/em><\/strong> users were disappointed, feeling the innovations were limited. Following the backlash, OpenAI<\/strong> was forced to add GPT-4o<\/strong> back to the options just one day after the launch.<\/p>\n\n\n\n

OpenAI’s strategic partner, Microsoft<\/a><\/em><\/strong>, also began turning to Anthropic’s models<\/strong> after GPT-5<\/strong> failed to meet expectations. Currently, Anthropic’s<\/strong> models are being used in Copilot Researcher<\/strong>, GitHub Copilot<\/strong>, Copilot Studio<\/strong>, and the new Office Agent<\/strong> systems capable of creating Word and PowerPoint documents.<\/p>\n\n\n\n

The GPT-5.1<\/strong> announcement follows OpenAI’s<\/strong> recent introduction of its AI-powered web browser, ChatGPT Atlas<\/strong>. This browser features an “agent mode<\/strong>,” available only to ChatGPT Plus<\/strong> and Pro<\/strong> users. This feature, similar to OpenAI’s Operator<\/strong> tool, can perform automated actions on the browser at the user’s request.<\/p>\n\n\n\n

You Might Also Like;<\/h3>\n\n\n