{"id":15021,"date":"2024-07-08T08:58:06","date_gmt":"2024-07-08T00:58:06","guid":{"rendered":"https:\/\/www.1ai.net\/?p=15021"},"modified":"2024-07-08T08:58:06","modified_gmt":"2024-07-08T00:58:06","slug":"%e4%b9%a6%e2%bd%a3%c2%b7%e6%b5%a6%e8%af%ad2-5-internlm2-5-7b%e6%a8%a1%e5%9e%8b%e5%ae%a3%e5%b8%83%e5%bc%80%e6%ba%90-%e6%94%af%e6%8c%81%e5%a4%84%e7%90%86%e7%99%be%e4%b8%87%e5%ad%97","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/15021.html","title":{"rendered":"Shusheng\u00b7Puyu 2.5\u2014\u2014InternLM2.5-7B model announced to be open source to support processing of million-word long articles"},"content":{"rendered":"

On July 3, 2024, Shanghai Artificial Intelligence Laboratory and SenseTime Technology, together with the Chinese University of Hong Kong and Fudan University, officially released a new generation ofLarge Language Model<\/a>Scholar Pu Yu 2.5 (InternLM2.5<\/a>).<\/p>\n

InternLM2.5-7B model has beenOpen Source<\/a>, and other scale models will be open sourced in due course. Shanghai Artificial Intelligence Laboratory promises to continue to provide free commercial licenses and support community innovation through high-quality open source models.<\/p>\n

\"\"<\/p>\n

The model has significant improvements in the following aspects:<\/p>\n

\n
    \n
  1. Reasoning ability: The reasoning ability of InternLM2.5 has been significantly enhanced, surpassing the Llama3-70B model in some dimensions. In particular, it has achieved a performance improvement of 100% on the mathematics evaluation set MATH, with an accuracy of 60%, which is comparable to the GPT-4Turbo1106 version.<\/li>\n
  2. Long text support: The model supports contexts of up to 1M tokens, which means it can process long texts of about 1.2 million Chinese characters. By increasing the context length and synthesizing data, it optimizes long document understanding and agent interaction.<\/li>\n
  3. Autonomous planning and tool calling: It has the ability to search and integrate information from hundreds of web pages. Through the MindSearch multi-agent framework, it simulates the human thinking process and effectively integrates network information.<\/li>\n<\/ol>\n<\/blockquote>\n

    Github Link<\/strong>:https:\/\/github.com\/InternLM\/InternLM<\/p>\n

    Model link:<\/strong>https:\/\/www.modelscope.cn\/models\/Shanghai_AI_Laboratory\/internlm2_5-7b-chat<\/p>\n

    Scholar Pu Yu's homepage:<\/strong>https:\/\/internlm.intern-ai.org.cn\/<\/p>","protected":false},"excerpt":{"rendered":"

    On July 3, 2024, Shanghai Artificial Intelligence Laboratory (AIL) and Shangtang Technology, in conjunction with The Chinese University of Hong Kong and Fudan University, officially released the next generation of big language model book \u2f63-\u6d66\u8bed2.5 (InternLM2.5). The InternLM2.5-7B model has been open-sourced, and other models of other scales will follow. Shanghai Artificial Intelligence Laboratory (SAL) is committed to continue to provide free commercial licenses and support community innovation through high-quality open source models. The model has significant improvements in the following aspects: Reasoning Ability:The reasoning ability of InternLM2.5 is significantly enhanced, surpassing the Llama3-70B model in some dimensions, and especially achieving a performance improvement of 100% on the mathematical review set MATH, with an accuracy of 60%, which is comparable to that of the GPT-4Turbo1106 version.<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[3395,706,219],"collection":[],"class_list":["post-15021","post","type-post","status-publish","format-standard","hentry","category-news","tag-internlm2-5","tag-706","tag-219"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15021","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=15021"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/15021\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=15021"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=15021"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=15021"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=15021"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}