{"id":13000,"date":"2025-08-25T12:16:28","date_gmt":"2025-08-25T03:16:28","guid":{"rendered":"https:\/\/aireviewirush.com\/?p=13000"},"modified":"2025-08-25T12:16:28","modified_gmt":"2025-08-25T03:16:28","slug":"the-newest-gemini-nano-with-on-device-ml-equipment-genai-apis","status":"publish","type":"post","link":"https:\/\/aireviewirush.com\/?p=13000","title":{"rendered":"The newest Gemini Nano with on-device ML Equipment GenAI APIs"},"content":{"rendered":"<p> <br \/>\n<\/p>\n<div>\n<div>\n<meta content=\"https:\/\/blogger.googleusercontent.com\/img\/b\/R29vZ2xl\/AVvXsEjx-b7W3hM_0hhBTzsSVx7R0fqmOY-7TGUiWj61mqPUvr83xNMsm1XjJ4XzQu8FCCMeno9tKLr3j2LlJ5lKtgLkMCOiwYzzhE6ZrCTQxbzZvEBAcAFulPR_yiute3DQty_itwOI6N9xuDmuPz6S2iT71KWuf84HAaYYdJV0R18oI17A02__3M_VmTNQXX4\/s1600\/Gemini-Nano-on-Android.png\" name=\"twitter:image\"\/><br \/>\n<img decoding=\"async\" src=\"https:\/\/blogger.googleusercontent.com\/img\/b\/R29vZ2xl\/AVvXsEjx-b7W3hM_0hhBTzsSVx7R0fqmOY-7TGUiWj61mqPUvr83xNMsm1XjJ4XzQu8FCCMeno9tKLr3j2LlJ5lKtgLkMCOiwYzzhE6ZrCTQxbzZvEBAcAFulPR_yiute3DQty_itwOI6N9xuDmuPz6S2iT71KWuf84HAaYYdJV0R18oI17A02__3M_VmTNQXX4\/s1600\/Gemini-Nano-on-Android.png\" alt=\"\"><\/p>\n<p><em>Posted by Caren Chang &#8211; Developer Relations Engineer, Joanna (Qiong) Huang &#8211; Software program Engineer, and Chengji Yan &#8211; Software program Engineer<\/em><\/p>\n<p><a href=\"https:\/\/blogger.googleusercontent.com\/img\/b\/R29vZ2xl\/AVvXsEj5Zj_BA7WbV_aES5kXV4y0MCKoVamBpNtHhFmBTkbUO0gQL545YeHqXNcx_j_YbdC1Lf2UZt7LhzXjGoVi1BB_8PzwTqMTSVFNmZpgOhSYuXqXbCi0XIBu9JJ6ZncnrudVNAY1M8OgGK7rvROJn0zCVHX1_f7rRT8k5YfNfMUipUvStWJDHl5c9xlMLkA\/s1600\/Android_Evergreen_Hero_Banner_AI_WebandApps.png\" target=\"_blank\" rel=\"noopener\"><img decoding=\"async\" border=\"0\" data-original-height=\"800\" data-original-width=\"100%\" src=\"https:\/\/blogger.googleusercontent.com\/img\/b\/R29vZ2xl\/AVvXsEj5Zj_BA7WbV_aES5kXV4y0MCKoVamBpNtHhFmBTkbUO0gQL545YeHqXNcx_j_YbdC1Lf2UZt7LhzXjGoVi1BB_8PzwTqMTSVFNmZpgOhSYuXqXbCi0XIBu9JJ6ZncnrudVNAY1M8OgGK7rvROJn0zCVHX1_f7rRT8k5YfNfMUipUvStWJDHl5c9xlMLkA\/s1600\/Android_Evergreen_Hero_Banner_AI_WebandApps.png\" alt=\"\"><\/a><\/p>\n<p>The newest model of Gemini Nano, our strongest multi-modal on-device mannequin, <a href=\"https:\/\/blog.google\/products\/pixel\/tensor-g5-pixel-10\/\" target=\"_blank\" rel=\"noopener\">simply launched on the Pixel 10 machine collection<\/a> and is now accessible by way of the <a href=\"https:\/\/developers.google.com\/ml-kit\/genai\" target=\"_blank\" rel=\"noopener\">ML Equipment GenAI APIs<\/a>. Combine capabilities equivalent to <a href=\"https:\/\/android-developers.googleblog.com\/2025\/05\/on-device-gen-ai-apis-ml-kit-gemini-nano.html\" target=\"_blank\" rel=\"noopener\">summarization, proofreading, rewriting, and picture description immediately into your apps<\/a>.<\/p>\n<p>With GenAI APIs we\u2019re centered on providing you with entry to the most recent model of Gemini Nano whereas offering constant high quality throughout gadgets and mannequin upgrades.  Right here\u2019s a sneak peak behind the scenes of among the issues we\u2019ve finished to attain this.<\/p>\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_53 counter-hierarchy ez-toc-counter ez-toc-grey ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title \" >Table of Contents<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\" role=\"button\"><label for=\"item-6a32a61245680\" ><span class=\"\"><span style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/label><input aria-label=\"Toggle\" aria-label=\"item-6a32a61245680\"  type=\"checkbox\" id=\"item-6a32a61245680\"><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/aireviewirush.com\/?p=13000\/#Adapting_GenAI_APIs_for_the_most_recent_Gemini_Nano\" title=\"Adapting GenAI APIs for the most recent Gemini Nano\">Adapting GenAI APIs for the most recent Gemini Nano<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/aireviewirush.com\/?p=13000\/#The_newest_Gemini_Nano_efficiency\" title=\"The newest Gemini Nano efficiency\">The newest Gemini Nano efficiency<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/aireviewirush.com\/?p=13000\/#The_way_forward_for_Gemini_Nano_with_GenAI_APIs\" title=\"The way forward for Gemini Nano with GenAI APIs\">The way forward for Gemini Nano with GenAI APIs<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/aireviewirush.com\/?p=13000\/#Be_taught_extra_about_GenAI_APIs\" title=\"Be taught extra about GenAI APIs\">Be taught extra about GenAI APIs<\/a><\/li><\/ul><\/nav><\/div>\n<h2><span class=\"ez-toc-section\" id=\"Adapting_GenAI_APIs_for_the_most_recent_Gemini_Nano\"><\/span><span style=\"font-size: x-large;\">Adapting GenAI APIs for the most recent Gemini Nano<\/span><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>We wish to make it as simple as doable so that you can construct AI powered options, utilizing probably the most highly effective fashions. To make sure GenAI APIs present constant high quality throughout completely different mannequin variations, we make many behind the scenes enhancements together with rigorous evals and adapter coaching.<\/p>\n<ol>\n<li><b>Analysis pipeline:<\/b> For every supported language, we put together an analysis dataset. We then benchmark the evals by way of a mixture of: LLM-based raters, statistical metrics and human raters.<\/li>\n<li><b>Adapter coaching:<\/b> With outcomes from the analysis pipeline, we then decide if we have to prepare feature-specific LoRA adapters to be deployed on prime of the Gemini Nano base mannequin. By transport GenAI APIs with LoRA adapters, we guarantee every API meets our high quality bar whatever the model of Gemini Nano working on a tool.<\/li>\n<\/ol>\n<h2><span class=\"ez-toc-section\" id=\"The_newest_Gemini_Nano_efficiency\"><\/span><span style=\"font-size: x-large;\">The newest Gemini Nano efficiency<\/span><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>One space we&#8217;re enthusiastic about is how this up to date model of Gemini Nano pushes efficiency even greater, particularly the prefix velocity &#8211; that&#8217;s how briskly the mannequin processes enter.<\/p>\n<p>For instance, listed below are outcomes when working text-to-text and image-to-text benchmarks on a Pixel 10 Professional.<\/p>\n<div align=\"left\">\n<table class=\"fr-table-selection-hover\" style=\"border-collapse: collapse; border: none; width: 100%;\">\n<tbody>\n<tr>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">\n                  <b>Prefix Velocity &#8211; Gemini <span style=\"color: #0d904f ; font-family: courier;\">nano-v2<\/span> on Pixel 9 Professional<\/b>\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">\n                  \t<b>Prefix Velocity &#8211; Gemini <span style=\"color: #0d904f ; font-family: courier;\">nano-v2<sup>*<\/sup><\/span> on Pixel 10 Professional<\/b>\n                \t<\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">\n                  \t<b>Prefix Velocity &#8211; Gemini <span style=\"color: #0d904f ; font-family: courier;\">nano-v3<\/span> on Pixel 10 Professional<\/b>\n                \t<\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">Textual content-to-text\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">510 tokens\/second\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">610 tokens\/second\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">940 tokens\/second\n                <\/td>\n<\/tr>\n<tr>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">Picture-to-text\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">510 tokens\/second + 0.8 seconds for picture encoding\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">610 tokens\/second + 0.7 seconds for picture encoding\n                <\/td>\n<td style=\"border: 1pt solid rgb(0, 0, 0); width: 25%;\">940 tokens\/second + 0.6 seconds for picture encoding\n                <\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p><center><sup>*<\/sup><small><em>Experimentation with Gemini <span style=\"color: #0d904f ; font-family: courier;\">nano-v2<\/span> on Pixel 10 Professional for benchmarking functions. All Pixel 10 Execs launched with Gemini <span style=\"color: #0d904f ; font-family: courier;\">nano-v3<\/span>.<\/em><\/small><\/center><\/p>\n<h2><span class=\"ez-toc-section\" id=\"The_way_forward_for_Gemini_Nano_with_GenAI_APIs\"><\/span><span style=\"font-size: x-large;\">The way forward for Gemini Nano with GenAI APIs<\/span><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>As we proceed to enhance the Gemini Nano mannequin, the staff is dedicated to utilizing the identical course of to make sure constant and prime quality outcomes from GenAI APIs.<\/p>\n<p>We hope it will considerably scale back the hassle to combine Gemini Nano in your Android apps whereas nonetheless permitting you to take full benefit of recent variations and their improved capabilites.<\/p>\n<h2><span class=\"ez-toc-section\" id=\"Be_taught_extra_about_GenAI_APIs\"><\/span><span style=\"font-size: x-large;\">Be taught extra about GenAI APIs<\/span><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Begin implementing GenAI APIs in your Android apps at the moment with steering from our official <a href=\"https:\/\/developers.google.com\/ml-kit\/genai\" target=\"_blank\" rel=\"noopener\">documentation<\/a> and samples: <a href=\"https:\/\/github.com\/android\/ai-samples\/tree\/main\/ai-catalog\/samples\" target=\"_blank\" rel=\"noopener\">GenAI API Catalog<\/a> and <a href=\"https:\/\/github.com\/googlesamples\/mlkit\/tree\/master\/android\/genai\" target=\"_blank\" rel=\"noopener\">ML Equipment GenAI APIs quickstart samples<\/a>.<\/p>\n<\/div>\n<hr\/>\n<\/div>\n\n","protected":false},"excerpt":{"rendered":"<p>Posted by Caren Chang &#8211; Developer Relations Engineer, Joanna (Qiong) Huang &#8211; Software program Engineer, and Chengji Yan &#8211; Software program Engineer The newest model of Gemini Nano, our strongest multi-modal on-device mannequin, simply launched on the Pixel 10 machine collection and is now accessible by way of the ML Equipment GenAI APIs. Combine capabilities [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":13002,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[23],"tags":[],"class_list":["post-13000","post","type-post","status-publish","format-standard","has-post-thumbnail","category-mobile"],"_links":{"self":[{"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/posts\/13000","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=13000"}],"version-history":[{"count":1,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/posts\/13000\/revisions"}],"predecessor-version":[{"id":13001,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/posts\/13000\/revisions\/13001"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=\/wp\/v2\/media\/13002"}],"wp:attachment":[{"href":"https:\/\/aireviewirush.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=13000"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=13000"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aireviewirush.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=13000"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}