{"id":370332,"date":"2026-06-17T08:52:00","date_gmt":"2026-06-17T15:52:00","guid":{"rendered":"https:\/\/cms-articles.softonic.io\/en\/?p=370332"},"modified":"2026-06-17T08:52:08","modified_gmt":"2026-06-17T15:52:08","slug":"nvidia-unveils-enpire-self-training-robots-reach-99-success","status":"publish","type":"post","link":"https:\/\/cms-articles.softonic.io\/en\/nvidia-unveils-enpire-self-training-robots-reach-99-success\/","title":{"rendered":"Nvidia unveils ENPIRE: self-training robots reach 99% success"},"content":{"rendered":"<p class=\"wp-block-paragraph\">Nvidia has unveiled <strong>ENPIRE<\/strong>, a research system built with Carnegie Mellon University and the University of California, Berkeley. The idea is simple enough: train robots on real hardware instead of only in simulation, and do it with very little human involvement.<\/p>\n\n<p class=\"wp-block-paragraph\">Nvidia says ENPIRE runs <strong>a fully autonomous loop<\/strong> from start to finish. It resets the workspace, tries a strategy, checks the result, and updates its policy.<\/p>\n\n<p class=\"wp-block-paragraph\">In the company&#8217;s testing, <strong>eight robots<\/strong> reached success rates as high as 99% on manipulation tasks that would normally need manual resets, hand-built reward functions, and close engineering oversight.<\/p>\n\n<p class=\"wp-block-paragraph\">Those tests covered Push-T, sorting pins into a box, cutting or fastening cable ties, pin insertion, and assembly-style work like installing a GPU into a motherboard.<\/p>\n\n<p class=\"wp-block-paragraph\">Nvidia also says ENPIRE can generate reward functions and success checks from only <strong>a few minutes<\/strong> of success-and-failure video. It does that by using visual alignment, gripper height, estimated force, and views from two cameras, while keeping response times under 150 ms.<\/p>\n\n<p class=\"wp-block-paragraph\">If you follow robotics, this one deserves a look.<\/p>\n\n<p class=\"wp-block-paragraph\">After a lightly supervised initial setup, Nvidia says <strong>eight dual-arm YAM stations<\/strong> can share code through <a href=\"https:\/\/git.en.softonic.com\/\" rel=\"noopener\">Git<\/a>, read papers, test hypotheses, and decide whether behavior cloning or reinforcement learning makes more sense.<\/p>\n\n<p class=\"wp-block-paragraph\">In one example, Nvidia says ENPIRE reached <strong>100% pin insertion<\/strong> faster than a human-in-the-loop baseline. The company says ENPIRE builds on Nvidia&#8217;s <a href=\"https:\/\/eurekalc.en.softonic.com\/\" rel=\"noopener\">Eureka<\/a>, which beat human-written rewards on more than 80% of tasks, with gains above 50%, and Nvidia plans to open-source ENPIRE alongside Project GR00T and Nvidia Isaac.<\/p>","protected":false},"excerpt":{"rendered":"<p>Nvidia has unveiled ENPIRE, a research system built with Carnegie Mellon University and the University of California, Berkeley. The idea is simple enough: train robots on real hardware instead of only in simulation, and do it with very little human involvement. Nvidia says ENPIRE runs a fully autonomous loop from start to finish. It resets &hellip; <a href=\"https:\/\/cms-articles.softonic.io\/en\/nvidia-unveils-enpire-self-training-robots-reach-99-success\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Nvidia unveils ENPIRE: self-training robots reach 99% success&#8221;<\/span><\/a><\/p>\n","protected":false},"author":9332,"featured_media":370331,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","wpcf-pageviews":0},"categories":[1015],"tags":[],"usertag":[],"vertical":[],"content-category":[6771],"class_list":["post-370332","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","content-category-ai"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/370332","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/users\/9332"}],"replies":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/comments?post=370332"}],"version-history":[{"count":1,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/370332\/revisions"}],"predecessor-version":[{"id":370333,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/posts\/370332\/revisions\/370333"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/media\/370331"}],"wp:attachment":[{"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/media?parent=370332"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/categories?post=370332"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/tags?post=370332"},{"taxonomy":"usertag","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/usertag?post=370332"},{"taxonomy":"vertical","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/vertical?post=370332"},{"taxonomy":"content-category","embeddable":true,"href":"https:\/\/cms-articles.softonic.io\/en\/wp-json\/wp\/v2\/content-category?post=370332"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}