{"id":95326,"date":"2026-03-18T11:11:17","date_gmt":"2026-03-18T08:11:17","guid":{"rendered":"https:\/\/forklog.com\/en\/?p=95326"},"modified":"2026-03-18T14:21:46","modified_gmt":"2026-03-18T11:21:46","slug":"openai-unveils-fast-ai-models-gpt-5-4-mini-and-gpt-5-4-nano","status":"publish","type":"post","link":"https:\/\/forklog.com\/en\/openai-unveils-fast-ai-models-gpt-5-4-mini-and-gpt-5-4-nano\/","title":{"rendered":"OpenAI Unveils Fast AI Models GPT-5.4 Mini and GPT-5.4 Nano"},"content":{"rendered":"<p>OpenAI has launched two new AI models: GPT-5.4 mini and GPT-5.4 nano. These are specifically optimized for tasks requiring minimal latency in response generation.<\/p>\n<blockquote class=\"twitter-tweet\">\n<p dir=\"ltr\" lang=\"en\">We\u2019re introducing GPT-5.4 mini and nano, our most capable small models yet.<\/p>\n<p>GPT-5.4 mini is more than 2x faster than GPT-5 mini. Optimized for coding, computer use, multimodal understanding, and subagents.<\/p>\n<p>For lighter-weight tasks, GPT-5.4 nano is our smallest and cheapest\u2026 <a href=\"https:\/\/t.co\/cdp5HWtM2M\">pic.twitter.com\/cdp5HWtM2M<\/a><\/p>\n<p>\u2014 OpenAI Developers (@OpenAIDevs) <a href=\"https:\/\/twitter.com\/OpenAIDevs\/status\/2033953815834333608?ref_src=twsrc%5Etfw\">March 17, 2026<\/a><\/p><\/blockquote>\n<p><script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<p>The firm stated that GPT\u20115.4 mini significantly surpasses GPT\u20115 mini in coding, logical reasoning, multimodal analysis, and working with third-party tools. The data processing speed has more than doubled.<\/p>\n<p>In several tests such as SWE-Bench Pro and OSWorld-Verified, the neural network approached the performance of GPT\u20115.4.<\/p>\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-276939\" src=\"https:\/\/forklog.com\/wp-content\/uploads\/img-fb72d40bf570b96d-9037592336771265.webp\" alt=\"image\" \/><figcaption class=\"wp-element-caption\">Source: OpenAI.<\/figcaption><\/figure>\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p>\u201cGPT\u20115.4 mini is designed for tasks where response speed directly determines product quality,\u201d the announcement states.<\/p><\/blockquote>\n<p>OpenAI outlined scenarios where the new solutions are most effective:<\/p>\n<ul class=\"wp-block-list\">\n<li>AI assistants with instant response;<\/li>\n<li>subagents that instantly perform auxiliary functions;<\/li>\n<li>interface management systems that create and interpret screenshots;<\/li>\n<li>multimodal applications for real-time image processing.<\/li>\n<\/ul>\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p>\u201cIn such scenarios, the best model is not necessarily the most powerful. It provides instant answers, works reliably with tools, and successfully handles complex professional tasks,\u201d the startup stated.<\/p><\/blockquote>\n<p>GPT\u20115.4 nano is the most compact and fastest version of GPT\u20115.4. The company recommends using it for classification and data extraction, ranking, and subagents performing simple auxiliary tasks.<\/p>\n<h2 class=\"wp-block-heading\">Programming<\/h2>\n<p>OpenAI emphasized that GPT\u20115.4 mini is particularly effective in development processes where high iteration speed is crucial. The model with minimal latency handles pinpoint editing, codebase navigation, frontend generation, and debugging cycles.<\/p>\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p>\u201cThis makes it an excellent choice for tasks requiring high speed and low costs,\u201d the announcement states.<\/p><\/blockquote>\n<p>In benchmarks, GPT\u20115.4 mini outperforms GPT\u20115 mini with comparable response speed, while operating significantly faster.<\/p>\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-276940\" src=\"https:\/\/forklog.com\/wp-content\/uploads\/img-98f36a5c614b03ed-9037659763815237-1024x572.png\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" srcset=\"https:\/\/forklog.com\/wp-content\/uploads\/img-98f36a5c614b03ed-9037659763815237-1024x572.png 1024w, https:\/\/forklog.com\/wp-content\/uploads\/img-98f36a5c614b03ed-9037659763815237-300x168.png 300w, https:\/\/forklog.com\/wp-content\/uploads\/img-98f36a5c614b03ed-9037659763815237-768x429.png 768w, https:\/\/forklog.com\/wp-content\/uploads\/img-98f36a5c614b03ed-9037659763815237.png 1217w\" alt=\"image\" width=\"1024\" height=\"572\" \/><figcaption class=\"wp-element-caption\">Source: OpenAI.<\/figcaption><\/figure>\n<h2 class=\"wp-block-heading\">Computer Use<\/h2>\n<p>GPT\u20115.4 mini demonstrates \u201cexcellent\u201d results in multimodal tasks related to interface management. The model can instantly interpret screenshots of overloaded user control panels for quick computer actions.<\/p>\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-276941\" src=\"https:\/\/forklog.com\/wp-content\/uploads\/img-d5efb4731682ff4b-9037677212732050.webp\" alt=\"image\" \/><figcaption class=\"wp-element-caption\">Source: OpenAI.<\/figcaption><\/figure>\n<h2 class=\"wp-block-heading\">Access and Pricing<\/h2>\n<p>GPT\u20115.4 mini is available via <span class=\"old_tooltip\" data-descr=\"application programming interface\">API<\/span>, Codex, and ChatGPT. Its usage costs $0.75 per million input tokens and $4.5 per million output tokens.<\/p>\n<p>GPT\u20115.4 nano is offered exclusively through API. Its rates are lower: $0.2 per million input tokens and $1.25 per million output tokens.<\/p>\n<p>In March, OpenAI <a href=\"https:\/\/forklog.com\/en\/news\/openai-integrates-shazam-into-chatgpt\">integrated<\/a> the music recognition service Shazam into ChatGPT. Users can identify tracks without leaving the application.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI has launched two new AI models: GPT-5.4 mini and GPT-5.4 nano. These are specifically optimized for tasks requiring minimal latency in response generation.<\/p>\n","protected":false},"author":1,"featured_media":95327,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"select":"1","news_style_id":"1","cryptorium_level":"","_short_excerpt_text":"OpenAI introduces GPT-5.4 mini and nano, optimized for minimal latency tasks.","creation_source":"ai_translated","_metatest_mainpost_news_update":false,"footnotes":""},"categories":[3],"tags":[438,1201,1190],"class_list":["post-95326","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-and-analysis","tag-artificial-intelligence","tag-chatbots","tag-openai"],"aioseo_notices":[],"amp_enabled":true,"views":"336","promo_type":"1","layout_type":"1","short_excerpt":"OpenAI introduces GPT-5.4 mini and nano, optimized for minimal latency tasks.","is_update":"0","_links":{"self":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/95326","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/comments?post=95326"}],"version-history":[{"count":2,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/95326\/revisions"}],"predecessor-version":[{"id":95343,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/95326\/revisions\/95343"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media\/95327"}],"wp:attachment":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media?parent=95326"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/categories?post=95326"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/tags?post=95326"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}