{"id":160985,"date":"2025-08-06T15:04:02","date_gmt":"2025-08-06T15:04:02","guid":{"rendered":"https:\/\/teknomers.com\/en\/how-to-install-openais-new-got-oss-models-on-your-computer-to-have-your-own-chatgpt-at-home\/"},"modified":"2025-08-06T15:04:02","modified_gmt":"2025-08-06T15:04:02","slug":"how-to-install-openais-new-got-oss-models-on-your-computer-to-have-your-own-chatgpt-at-home","status":"publish","type":"post","link":"https:\/\/teknomers.com\/en\/how-to-install-openais-new-got-oss-models-on-your-computer-to-have-your-own-chatgpt-at-home\/","title":{"rendered":"How to Install OpenAI&#8217;s New Got-Oss Models on Your Computer to Have Your Own ChatGPT at Home"},"content":{"rendered":"\n<p>OpenAI has announced new models that are <em>Open Source<\/em> and available for anyone to download and install on their own computers: \u00a0GPT-Oss\u00a0. With these models now released, it&#8217;s the perfect time to implement AI directly within your premises, in other words, \u00a0executing it on your computer\u00a0. Today, we will guide you through the steps to install and use these models effectively.<\/p>\n<p><!-- BREAK 1 --><\/p>\n<h2>Differences Between the Two Models<\/h2>\n<p>Even though they share a similar name, \u00a0GPT-Oss-120B and GPT-Oss-20B\u00a0 are distinctly different and come with varying requirements. The primary model, \u00a0GPT-Oss-120B\u00a0, is comparable to OpenAI\u2019s O4-mini, demanding at least \u00a060 GB of graphics memory\u00a0.<\/p>\n<p><!-- BREAK 2 --><\/p>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"632\" width=\"802\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/f85361\/captura-de_pantalla-2025-08-06-131717\/450_1000.png\" alt=\"Having your own ChatGPT at home is easy, but it requires a capable team | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"Having your own ChatGPT at home is easy, but it requires a capable team | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/f85361\/captura-de_pantalla-2025-08-06-131717\/450_1000.png\"\/><br \/>\n        <span>Having your own ChatGPT at home is easy, but it requires a capable team | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>The smaller counterpart, \u00a0GPT-Oss-20B\u00a0, is less powerful, akin to O3-mini according to OpenAI, and can be run on \u00a0edge devices\u00a0. This means it can operate on your personal computer, provided it meets the minimum requirement of \u00a016 GB of memory\u00a0, preferably graphics memory.<\/p>\n<p><!-- BREAK 3 --><\/p>\n<p>In summary:<\/p>\n<ul>\n<li><strong>GPT-Oss-120B<\/strong>: A large model needing at least 60 GB of VRAM or unified memory and is not suited for consumer computers.<\/li>\n<li><strong>GPT-Oss-20B<\/strong>: A smaller model requiring 16 GB of VRAM or unified memory, ideal for consumer computers.<\/li>\n<\/ul>\n<p>For this guide, we will focus on using \u00a0GPT-Oss-20B\u00a0.<\/p>\n<h2>Considerations to Take Into Account<\/h2>\n<p>Running AI locally is an \u00a0intensive process\u00a0, and may cause your computer to \u00a0slow down significantly\u00a0. While it is technically possible to execute the model with only 16 GB of RAM, it is recommended to have a \u00a0high-end GPU\u00a0 for optimal performance.<\/p>\n<p><!-- BREAK 4 --><\/p>\n<p>If your computer has less than 16 GB of VRAM, the tool will resort to using RAM, which should also be equal to or greater than 16 GB. Otherwise, the system may not function correctly. As a general piece of advice, it&#8217;s ideal to allocate as many resources as possible to the model\u2019s execution, so it is wise to close any unnecessary applications.<\/p>\n<p><!-- BREAK 5 --><\/p>\n<h2>Installing Ollama on Your Computer<\/h2>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"663\" width=\"897\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/2db4be\/captura-de_pantalla-2025-08-06-124555\/450_1000.png\" alt=\"OLLAMA Installation | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"OLLAMA Installation | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/2db4be\/captura-de_pantalla-2025-08-06-124555\/450_1000.png\"\/><br \/>\n        <span>OLLAMA Installation | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>For this tutorial, we will use a well-established application: \u00a0Ollama\u00a0. This Open Source platform significantly simplifies the installation, access, and use of \u00a0LLMs (Large Language Models)\u00a0. Think of it as a model executor.<\/p>\n<p><!-- BREAK 6 --><\/p>\n<p>ChatGPT operates online and allows users to interact with a model like GPT-4O, while Ollama provides the same capabilities, \u00a0but at home using models installed on your computer\u00a0. It is free, open-source software available for <a rel=\"noopener, noreferrer nofollow\" href=\"https:\/\/ollama.com\/download\/windows\" target=\"_blank\">Windows<\/a>, <a rel=\"noopener, noreferrer nofollow\" href=\"https:\/\/ollama.com\/download\/mac\" target=\"_blank\">Mac<\/a>, and <a rel=\"noopener, noreferrer nofollow\" href=\"https:\/\/ollama.com\/download\/linux\" target=\"_blank\">Linux<\/a>.<\/p>\n<p><!-- BREAK 7 --><\/p>\n<h2>Download GPT-Oss<\/h2>\n<p>Once Ollama is downloaded and installed on your computer, you\u2019ll encounter an interface resembling this. Alternatively, you can also opt for the terminal interface, but the graphical interface is significantly more user-friendly.<\/p>\n<p><!-- BREAK 8 --><\/p>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"632\" width=\"802\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/e06082\/captura-de_pantalla-2025-08-06-124744\/450_1000.png\" alt=\"Main interface of Ollama | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"Main interface of Ollama | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/e06082\/captura-de_pantalla-2025-08-06-124744\/450_1000.png\"\/><br \/>\n        <span>Main interface of Ollama | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>In the lower right corner, you&#8217;ll find a dropdown with the name of the model you&#8217;re currently using or will be using.<\/p>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"632\" width=\"802\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/abf7e9\/captura-de_pantalla-2025-08-06-124735\/450_1000.png\" alt=\"Access to the different AI models from Ollama | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"Access to the different AI models from Ollama | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/abf7e9\/captura-de_pantalla-2025-08-06-124735\/450_1000.png\"\/><br \/>\n        <span>Access to the different AI models from Ollama | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>By clicking on the dropdown, you&#8217;ll access several models, including Deepseek R1, Gemma, or Qwen. Here, we want to select &#8220;<strong>GPT-Oss: 20B<\/strong>&#8220;.<\/p>\n<p><!-- BREAK 9 --><\/p>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"632\" width=\"802\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/d47d27\/captura-de_pantalla-2025-08-06-125324\/450_1000.png\" alt=\"Download of the model, arm yourself with patience | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"Download of the model, arm yourself with patience | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/d47d27\/captura-de_pantalla-2025-08-06-125324\/450_1000.png\"\/><br \/>\n        <span>Download of the model, arm yourself with patience | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>After selecting &#8220;GPT-Oss: 20B,&#8221; just send a message in the chat to initiate the model&#8217;s download. Be patient because the model size is \u00a012.8 GB\u00a0, and downloading may take some time.<\/p>\n<p><!-- BREAK 10 --><\/p>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"632\" width=\"802\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/18bf75\/captura-de_pantalla-2025-08-06-131408\/450_1000.png\" alt=\"Talking to GPT-Oss-20B through Ollama | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"Talking to GPT-Oss-20B through Ollama | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/18bf75\/captura-de_pantalla-2025-08-06-131408\/450_1000.png\"\/><br \/>\n        <span>Talking to GPT-Oss-20B through Ollama | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>Once installed, you can start conversing with the AI just as you would with ChatGPT. However, if your GPU doesn&#8217;t meet the minimum requirements, you may notice that it operates much slower than ChatGPT. This difference is due to the fact that you&#8217;re executing the model on your own machine rather than accessing a powerful data center filled with the latest Nvidia dedicated GPUs.<\/p>\n<p><!-- BREAK 11 --><\/p>\n<h2>Another Option: LM Studio<\/h2>\n<div class=\"article-asset-image article-asset-normal article-asset-center\">\n<div class=\"asset-content\">\n<div class=\"caption-img \">\n                   <img class=\"centro_sinmarco\" height=\"760\" width=\"1200\" loading=\"lazy\" decoding=\"async\"  fetchpriority=\"high\"  src=\"https:\/\/i.blogs.es\/1f3333\/captura-de_pantalla-2025-08-06-140113\/450_1000.png\" alt=\"LM Studio | Image: Xataka\"\/><br \/>\n   <img decoding=\"async\" alt=\"LM Studio | Image: Xataka\" class=\"centro_sinmarco\" src=\"https:\/\/i.blogs.es\/1f3333\/captura-de_pantalla-2025-08-06-140113\/450_1000.png\"\/><br \/>\n        <span>LM Studio | Image: Xataka<\/span>\n   <\/div>\n<\/p><\/div>\n<\/div>\n<p>While Ollama is intuitive and straightforward, another comprehensive program you might consider is \u00a0[LM Studio](https:\/\/lmstudio.ai\/download)\u00a0. This application is accessible for Windows, Linux, and Mac, allowing it to manage multiple models, including \u00a0GPT-Oss: 20B\u00a0.<\/p>\n<p><!-- BREAK 12 --><\/p>\n<p>LM Studio provides more advanced options for fine-tuning both your computer&#8217;s settings and the model&#8217;s behavior, but fully leveraging it may require more advanced knowledge.<\/p>\n<p><!-- BREAK 13 --><\/p>\n<p>In conclusion, with the introduction of GPT-Oss models from OpenAI, you can now harness the power of advanced AI directly on your computer. Whether you choose Ollama for its simplicity or LM Studio for its expansive capabilities, both offer exciting opportunities to engage with AI in new and personalized ways.<\/p>\n<p>Cover image | Xataka<\/p>\n<p>In Xataka | How to move from image to video using artificial intelligence: 14 essential free tools<\/p>\n<p><br \/>\n<br \/><a href=\"https:\/\/teknomers.com\/category\/general\/\" rel=\"dofollow\">General News &#8211; 2<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI has announced new models that are Open Source and available for anyone to download and install on their own computers: \u00a0GPT-Oss\u00a0. With these models now released, it&#8217;s the perfect time to implement AI directly within your premises, in other words, \u00a0executing it on your computer\u00a0. Today, we will guide you through the steps to [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[36399],"tags":[12522,3532,40103,1134,39325,9859,27332],"class_list":["post-160985","post","type-post","status-publish","format-standard","hentry","category-technology","tag-chatgpt","tag-computer","tag-gotoss","tag-home","tag-install","tag-models","tag-openais"],"_links":{"self":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/160985","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/comments?post=160985"}],"version-history":[{"count":0,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/160985\/revisions"}],"wp:attachment":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media?parent=160985"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/categories?post=160985"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/tags?post=160985"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}