Understanding the Market | MEITUAN rises over 6% again, with a cumulative increase of over 20% this month. Citigroup points out that LongCat-2.0 open source is beneficial for competitive moats
Meituan-W rose over 6% again, with a cumulative increase of over 20% this month. As of the time of writing, it rose 6.19% to HKD 84.1, with a transaction volume of HKD 3.354 billion. In terms of news, on July 6, Meituan announced the official open-sourcing of all model weights, inference engines, and core technical documents for LongCat-2.0. This model is the first in the industry to complete full-process training and inference on a 50,000-card domestic computing cluster, with a total of 1.6 trillion parameters and a pre-training data scale exceeding 30 trillion tokens, natively supporting 1 million ultra-long contexts. Citigroup published a research report stating that Meituan's open-sourcing of LongCat-2.0 is expected to further consolidate its leadership position in the local life services market. The advanced Agentic architecture of the model will enhance internal R&D efficiency and attract external developers. The report indicated that by leveraging its vast and proprietary offline transaction and operational data, Meituan can provide unique AI-driven marketing and business insights for small and medium-sized enterprises. The bank believes this will create a competitive moat, distinguishing it from other large language model providers and deepening relationships with merchants and consumers
According to Zhitong Finance APP, Meituan-W (03690) has risen over 6% again, with a cumulative increase of over 20% this month. As of the time of writing, it is up 6.19%, priced at HKD 84.1, with a transaction volume of HKD 3.354 billion.
In terms of news, on July 6, Meituan announced the official open-sourcing of all model weights, inference engines, and core technical documents for LongCat-2.0. This model is the first in the industry to complete full-process training and inference on a domestic computing cluster with 50,000 cards, featuring a trillion-parameter large model with a total of 1.6T parameters and a pre-training data scale exceeding 30T tokens, natively supporting 1M ultra-long context.
Citigroup published a research report stating that Meituan's open-sourcing of LongCat-2.0 is expected to further consolidate its leadership position in the local life services market. The advanced Agentic architecture of this model will enhance internal R&D efficiency and attract external developers. The report indicates that by leveraging its vast and proprietary offline transaction and operational data, Meituan can provide unique AI-driven marketing and business insights for small and medium-sized enterprise merchants. The bank believes this will create a competitive moat, distinguishing it from other large language model providers and deepening relationships with merchants and consumers
