{"id":5464,"date":"2026-05-14T00:25:00","date_gmt":"2026-05-13T22:25:00","guid":{"rendered":"http:\/\/stocks-future.com\/?guid=62ea97564b1a032f44773b987d0d6614"},"modified":"2026-05-14T00:25:00","modified_gmt":"2026-05-13T22:25:00","slug":"cosimo-launches-its-physics-engine-encoding-a-new-kind-of-video-primitive-geometric-video-purpose-built-for-ai-reproducible-benchmarks-show-categorical-gains-over-legacy-video","status":"publish","type":"post","link":"https:\/\/stocks-future.com\/?p=5464","title":{"rendered":"COSIMO Launches its Physics Engine, Encoding a New Kind of Video Primitive, Geometric Video, Purpose-Built for AI. Reproducible Benchmarks Show Categorical Gains over Legacy Video"},"content":{"rendered":"<p>IRVINE, Calif.--(BUSINESS WIRE)--<a  href=\"https:\/\/cts.businesswire.com\/ct\/CT?id=smartlink&amp;url=http%3A%2F%2Fcosimo.ai&amp;esheet=54535692&amp;newsitemid=20260513526696&amp;lan=en-US&amp;anchor=COSIMO.AI&amp;index=1&amp;md5=2feb2808d5808d1ed5924c1e9ddb54ba\" rel=\"nofollow\" shape=\"rect\">COSIMO.AI<\/a> launches \u201cGeometric Video\u201d for AI, answering an industry problem: Legacy Video was never made for AI; it was made for human eyes.<\/p><br\/><a href=\"https:\/\/mms.businesswire.com\/media\/20260513526696\/en\/2804833\/5\/COSIMO-blue-transparent.jpg\"><img src=\"https:\/\/mms.businesswire.com\/media\/20260513526696\/en\/2804833\/22\/COSIMO-blue-transparent.jpg\" \/><\/a><br\/><a href=\"https:\/\/mms.businesswire.com\/media\/20260513526696\/en\/2804833\/5\/COSIMO-blue-transparent.jpg\"><img src=\"https:\/\/mms.businesswire.com\/media\/20260513526696\/en\/2804833\/21\/COSIMO-blue-transparent.jpg\" \/><\/a><p>\nThe Physical AI industry has been reduced to an anecdote. A robotaxi launch date slips. A humanoid demo gets delayed. The headlines predictably lament: \"AI is not ready.\"<\/p><p>\nThe industry's answer is always the same. Bigger models. More data. More GPUs. More data centers. More power. Hundreds of billions of dollars in, launch dates and demos keep slipping.<\/p><p>\nThe working assumption is that AI's efficiency is fixed and that only brute-force scaling will enable AI to accurately perceive video, the primary visual input of the physical world.<\/p><p>\nCOSIMO\u2019s <b>Physics Engine<\/b> answers this problem. It encodes the geometry of shapes and motion directly into the video stream. <b>The output is Geometric Video,<\/b> which captures the naked geometry of the world, strips out the noise, and encodes objects and their motion in a deterministic, mathematically pure form.<\/p><p>\nCOSIMO tested its Geometric Video on UCF-101. Five seeds. Forty epochs. NVIDIA L4 hardware. Against the Legacy Video baseline, Geometric Video delivers:<\/p><ul class=\"bwlistdisc\">\n<li>\n<b>+12.4 percentage points more accurate<\/b> than legacy video.<\/li>\n<li>\n<b>78.5% fewer model parameters.<\/b><\/li>\n<li>\n<b>27\u00d7 less GPU memory<\/b> at inference.<\/li>\n<li>\n<b>1.17 milliseconds<\/b> per frame on a five-year-old MacBook Pro. Under one watt.<\/li>\n<li>\n<b>3\u00d7 tighter accuracy clustering<\/b> across all five training runs. Stable enough to debug like source code.<\/li>\n<\/ul><p>\nEvery performance result traces to a public test run that has been cryptographically verified. The validation pipeline is available at <a  href=\"https:\/\/cts.businesswire.com\/ct\/CT?id=smartlink&amp;url=http%3A%2F%2Fcosimo.ai%2Fvalidation&amp;esheet=54535692&amp;newsitemid=20260513526696&amp;lan=en-US&amp;anchor=cosimo.ai%2Fvalidation&amp;index=2&amp;md5=5b385f0c29a844f397777198d1e17247\" rel=\"nofollow\" shape=\"rect\">cosimo.ai\/validation<\/a>.<\/p><p>\n<b>The economic implications are direct.<\/b> A Tier-1 Physical AI footprint saves $8 to $10 billion a year across compute, storage, bandwidth, and power, and saves $2,700 per edge device. The biggest savings are in time, where COSIMO\u2019s tighter accuracy clustering accelerates time-to-market schedules by an estimated 6 to 12 months. For a robotaxi or a humanoid program, that is a categorical advantage.<\/p><p>\nThe open-source savings calculator is at <a  href=\"https:\/\/cts.businesswire.com\/ct\/CT?id=smartlink&amp;url=http%3A%2F%2Fcosimo.ai%2Fsavings&amp;esheet=54535692&amp;newsitemid=20260513526696&amp;lan=en-US&amp;anchor=cosimo.ai%2Fsavings&amp;index=3&amp;md5=c45f11240644ef30c2fc085bf09d131b\" rel=\"nofollow\" shape=\"rect\">cosimo.ai\/savings<\/a>.<\/p><p>\nThe whitepaper is at: <a  href=\"https:\/\/cts.businesswire.com\/ct\/CT?id=smartlink&amp;url=http%3A%2F%2Fcosimo.ai%2Fwhitepaper&amp;esheet=54535692&amp;newsitemid=20260513526696&amp;lan=en-US&amp;anchor=cosimo.ai%2Fwhitepaper&amp;index=4&amp;md5=e1c9b166c59ffa4ebc3ff1548a5b697c\" rel=\"nofollow\" shape=\"rect\">cosimo.ai\/whitepaper<\/a>.<\/p><p>\n<b>About COSIMO.<\/b> COSIMO's Physics Engine encodes the geometry of shapes and motion directly into the video stream, generating \"Geometric Video,\" a new kind of video primitive, purpose-built for AI. <a  href=\"https:\/\/cts.businesswire.com\/ct\/CT?id=smartlink&amp;url=http%3A%2F%2Fwww.cosimo.ai&amp;esheet=54535692&amp;newsitemid=20260513526696&amp;lan=en-US&amp;anchor=www.cosimo.ai&amp;index=5&amp;md5=c923fc2d58c234942ada6bd893862637\" rel=\"nofollow\" shape=\"rect\">www.cosimo.ai<\/a><\/p><br\/> <b>Contacts<\/b> <br\/><p>\nMedia Contact: <a  href=\"mailto:press@cosimo.ai\" rel=\"nofollow\" shape=\"rect\">press@cosimo.ai<\/a><\/p>","protected":false},"excerpt":{"rendered":"<p>IRVINE, Calif.&#8211;(BUSINESS WIRE)&#8211;COSIMO.AI launches \u201cGeometric Video\u201d for AI, answering an industry problem: Legacy Video was never made for AI; it was made for human eyes.<br \/>\nThe Physical AI industry has been reduced to an anecdote. A robotaxi launch dat&#8230;<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-5464","post","type-post","status-publish","format-standard","hentry","category-infos-businesswire"],"_links":{"self":[{"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/posts\/5464","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/stocks-future.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5464"}],"version-history":[{"count":1,"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/posts\/5464\/revisions"}],"predecessor-version":[{"id":5465,"href":"https:\/\/stocks-future.com\/index.php?rest_route=\/wp\/v2\/posts\/5464\/revisions\/5465"}],"wp:attachment":[{"href":"https:\/\/stocks-future.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5464"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/stocks-future.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5464"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/stocks-future.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5464"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}