{"id":32569,"date":"2025-04-07T23:19:03","date_gmt":"2025-04-07T15:19:03","guid":{"rendered":"https:\/\/www.1ai.net\/?p=32569"},"modified":"2025-04-07T23:19:03","modified_gmt":"2025-04-07T15:19:03","slug":"%e4%b8%ad%e5%9b%bd%e7%a7%91%e5%ad%a6%e9%99%a2%e9%9d%92%e8%97%8f%e9%ab%98%e5%8e%9f%e7%a0%94%e7%a9%b6%e6%89%80%e3%80%81%e9%98%bf%e9%87%8c%e4%ba%91%e8%81%94%e5%90%88%e5%8f%91%e5%b8%83%e6%b0%b4%e8%83%bd","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/32569.html","title":{"rendered":"Tibetan Plateau Institute of Chinese Academy of Sciences and Aliyun jointly release \"Luoshu\", a multimodal inference model for water energy grain"},"content":{"rendered":"

April 7 News.Alibaba Cloud<\/a>An official release today announced that the recentChinese Academy of Sciences<\/a>Qinghai-Tibetan Plateau Research Institute<\/a>Jointly released with AliCloudhydroelectric foodA Large Model of Multimodal Reasoning<\/a>\u201cBook of Songs<\/a>\u201d<\/strong>.<\/p>\n

According to the introduction, the \"Luoshu\" large model integrates the scientific model \"SiYuan\" (Hydro Trace) and a thousand questions reasoning model QwQ-32B, as well as the multimodal model Qwen2.5-VL, which can accurately analyze and predict the amount and source of water coming from a specific region in different time scales, thus helping to regulate the dynamic balance of water supply and food production demand in water resources management. Qwen2.5-VL is a multimodal model that accurately analyzes and predicts the amount and source of water coming to a specific region at different time scales, thus helping to regulate the dynamic balance of water supply, power generation and food production needs in water resources management. CurrentlyThe prediction accuracy of \"Luoshu\" has reached 98%.<\/strong>(SOTA level), and will be gradually applied to several energy scenarios in the future.<\/p>\n

In addition, the bottom layer of \"Luoshu\" is a scientific model developed by the Qinghai-Tibet Institute of the Chinese Academy of Sciences (HydroTrace), relying on the spatial and temporal multimodal data training, the output contains two parts, one is the runoff volume that directly supports the prediction of hydropower production, and the other is the high-dimensional data that accurately depicts hydrological processes, but these data can not be directly understood and used by human beings. directly understood and used by humans.<\/p>\n

After \"Luoshu\" accesses QwQ-32B inference model and Qwen2.5-VL-32B multimodal model, users can reason and analyze and visualize these high-dimensional data by combining them with business logic using only natural language, and quantitatively interpret the composite heat map which is difficult to \"see through\" by the naked eye. \"The composite heat map, which is difficult to see with the naked eye, can be quantitatively interpreted, providing scientific and technological support for the integrated spatial and temporal scheduling of regional energy.<\/p>\n

\"8b4d0f4ej00suct6o009td000hs00dcm\"<\/p>\n

\"The training of the scientific model of Luoshu was supported by the \"Cloudworks\" program.Rely on AliCloud ECS to complete the whole process<\/strong>The \"Luoshu\" big model integrates the scientific big model and the multimodal big model of reasoning, and calls the Tongyi big model through the AliCloud Bailian platform. At present, the \"Luoshu\" model, which integrates scientific, reasoning and multimodal models, has also been launched on AliCloud's Bailian platform. Compared with traditional prediction methods, the accuracy of \"Luoshu\" has increased by nearly 20%, providing an innovative solution to the volatility of water and energy supply under climate change.<\/p>\n

1AI was officially informed by AliCloud that now \"Luoshu\" has completed two rounds of internal testing in the regional hydropower leading enterprises, water staff only need to open the \"Luoshu\" through the cell phone or computer, you can conveniently and efficiently analyze the hydrology, meteorology, climate and other multi-dimensional data, integrate water conditions and other resources, optimize the whole chain of resource allocation, cross-domain collaborative scheduling. Integration of water, electricity and other resources to optimize the allocation of resources across the chain, cross-domain coordinated scheduling.<\/p>","protected":false},"excerpt":{"rendered":"

On 7 April, the official Ariyun announced in an official communication today that, in recent days, the Chinese Academy of Sciences, in association with Ariyun, published a large model of hydro-food multimodular reasoning \u201cLo Book\u201d. It was described that the large \u201cLoss\u201d model, which was a collection of scientific models, Hydro Trace, and QwQ-32B, and the multi-mode model Qwen2.5-VL, allowed for precision analysis and prediction of water availability and sources in a given region at different time scales, thus helping to regulate the dynamic balance between water supply, power generation and food production needs in water resources management. At present, the forecast accuracy rate for \u201cLo\u201d is 98% (SOTA level) and will be gradually applied to multiple energy scenarios in the future. In addition, the bottom of Los Angeles is a self-researched science at the Quizhi Institute<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[4192,6208,6206,334,6207],"collection":[],"class_list":["post-32569","post","type-post","status-publish","format-standard","hentry","category-news","tag-4192","tag-6208","tag-6206","tag-334","tag-6207"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/32569","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=32569"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/32569\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=32569"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=32569"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=32569"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=32569"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}