As China attempts to move beyond using domestic chips solely for model inference, food delivery giant Meituan released what it claims is the country\u2019s largest artificial intelligence model trained entirely on home-grown hardware.<\/p>\n
The Beijing-based on-demand service giant on Tuesday open-sourced LongCat-2.0, a new large language model (LLM) boasting 1.6 trillion parameters and a context window of 1 million tokens. The scale puts it on par with DeepSeek\u2019s latest flagship model, V4-pro, which launched in April.<\/p>\n
Meituan claimed that LongCat-2.0 was the industry\u2019s first trillion-parameter model to complete full-process training and inference on a 50,000-card domestic computing power cluster.<\/p>\n
While DeepSeek-V4-pro relied on home-grown chips only for inference \u2013 the process where a pre-trained model runs to answer user queries \u2013 LongCat-2.0 used domestic hardware for both inference and pre-training, according to Meituan.<\/p>\n
Pre-training is a far more computationally intensive process, during which an AI model digests massive data sets to learn basic patterns.<\/p>\n
Meituan said LongCat-2.0 was built entirely on \u201clarge-scale clusters of tens of thousands of AI ASIC superpods\u201d, showing its ability to \u201cconduct frontier-scale training on alternative hardware platforms\u201d. An ASIC, or application-specific integrated circuit, is a chip customised for specific workloads, as opposed to a general-purpose processor.<\/p>\n","protected":false},"excerpt":{"rendered":"As China attempts to move beyond using domestic chips solely for model inference, food delivery giant Meituan released…\n","protected":false},"author":2,"featured_media":525454,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,3673,11702,218,219,223778,2102,114,4067,1711,223779,6160,61,60,223780,46703,233,1682,80,223781,75988],"class_list":["post-525453","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-anthropic","tag-artificial-analysis","tag-artificial-intelligence","tag-artificialintelligence","tag-atlas-950-superpods","tag-beijing","tag-china","tag-deepseek","tag-google","tag-huawei-collective-communication-library","tag-huawei-technologies","tag-ie","tag-ireland","tag-longcat-2-0","tag-meituan","tag-nvidia","tag-openai","tag-technology","tag-tp-huang","tag-zhipu-ai"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/525453","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=525453"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/525453\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/525454"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=525453"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=525453"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=525453"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}