{"id":15966,"date":"2024-07-20T08:21:33","date_gmt":"2024-07-20T00:21:33","guid":{"rendered":"https:\/\/www.1ai.net\/?p=15966"},"modified":"2024-07-20T08:21:33","modified_gmt":"2024-07-20T00:21:33","slug":"%e5%95%86%e6%b1%a4%e7%bb%9d%e5%bd%b1%e8%a1%8c%e4%b8%9a%e9%a6%96%e5%8f%91%e5%8e%9f%e7%94%9f%e5%a4%9a%e6%a8%a1%e6%80%81%e5%a4%a7%e6%a8%a1%e5%9e%8b%e8%bd%a6%e7%ab%af%e9%83%a8%e7%bd%b2%ef%bc%9a80","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/15966.html","title":{"rendered":"SenseTime Jueying launches the industry&#039;s first native multi-modal large model vehicle-side deployment: 8 billion parameters, 40 tokens per second"},"content":{"rendered":"<p><a href=\"https:\/\/www.1ai.net\/en\/tag\/%e5%95%86%e6%b1%a4%e7%a7%91%e6%8a%80\" title=\"[View articles tagged with [quotidian technology]]\" target=\"_blank\" >SenseTime<\/a>Co-founder and chief scientist Wang Xiaogang announced on the 17th,<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e5%95%86%e6%b1%a4%e7%bb%9d%e5%bd%b1\" title=\"&quot;Look at the article that contains the labels.&quot;\" target=\"_blank\" >Shang Tang Jue Ying<\/a>The first in the industry to achieve native<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e5%a4%9a%e6%a8%a1%e6%80%81%e5%a4%a7%e6%a8%a1%e5%9e%8b\" title=\"[Sees articles with [Multimodal Large Model] labels]\" target=\"_blank\" >Multimodal large model<\/a>The vehicle-side 8B model has a first packet delay of less than 300 milliseconds, an inference speed of 40 Tokens\/second, and covers mainstream computing platforms.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-15967\" title=\"19173ed2-ef7c-4472-8ca9-a98d856b4206.jpg@s_2w_820h_289\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/07\/19173ed2-ef7c-4472-8ca9-a98d856b4206.jpg@s_2w_820h_289.jpg\" alt=\"19173ed2-ef7c-4472-8ca9-a98d856b4206.jpg@s_2w_820h_289\" width=\"820\" height=\"289\" \/><\/p>\n<p>SenseTime has created a computing engine called &quot;HyperPPL&quot; for multimodal large models. It currently expands and supports mainstream in-vehicle computing hardware, is compatible with a variety of mainstream operating systems, and adapts to the deployment platforms of multiple in-vehicle chips.<\/p>\n<p>SenseTime Jueying said that HyperPPL is optimized for multi-person scenarios in vehicles, so that when there are multiple people in the vehicle at the same time, the model reasoning efficiency of the multi-modal large model on the vehicle side is not significantly reduced compared to a single person.<\/p>\n<p>SenseTime previously stated that Shenzhen\u2019s first autonomous driving bus line uses its vehicles and technology, and all driving operations do not require human intervention.<\/p>\n<p>Next year, automotive chips (NVIDIA Thor) with a computing power of over 1,000 TOPS will be available. Based on a computing platform with higher computing power, SenseTime expects that the first packet delay of Jueying&#039;s multi-modal large model vehicle-side deployment solution will be greatly reduced, and the inference speed will be further improved.<\/p>","protected":false},"excerpt":{"rendered":"<p>Wang Xiaogang, co-founder and chief scientist of Shangtang Technology, announced on the 17th that Shangtang Jade is the first in the industry to realize the vehicle-side deployment of native multimodal large models. The first packet latency of the 8B model on the vehicle side is less than 300 milliseconds, and the inference speed is 40 Tokens\/second, covering the mainstream computing power platforms. The computation engine \"HyperPPL\" for multimodal large models is currently expanding and supporting mainstream in-vehicle computing hardware, compatible with a variety of mainstream operating systems, and adapted to multiple in-vehicle chip deployment platforms. Business soup said HyperPPL optimized for the vehicle multi-person scenarios, so that the car concurrent multi-person situation, the car end of the multi-modal model model reasoning efficiency compared to a single person is not significantly reduced. Shangtang Jiyi said previously, Shenzhen's first automatic driving<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[2171,3605,602],"collection":[],"class_list":["post-15966","post","type-post","status-publish","format-standard","hentry","category-news","tag-2171","tag-3605","tag-602"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15966","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=15966"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15966\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=15966"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=15966"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=15966"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=15966"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}