88Z AI, 30 June — Meituan has officially open-sourced LongCat-2.0, a large model with 1.6 trillion total parameters and 48 billion average activated parameters. It is the first model in the industry to complete full-process training and inference on a domestic computing cluster of 50,000 cards, natively supporting 1M long context and optimized for Agentic Coding tasks.
At a glance
- LongCat-2.0 is the first model to complete full-process training and inference on a 50,000-card domestic computing cluster
- The model natively supports 1M context and is optimized for Agentic Coding, with its preview version ranking top globally in call volume
The update
Meituan recently announced the official open-source release of its new generation trillion-parameter large model, LongCat-2.0. The model features a total parameter count of 1.6T, with an average of 48B activated parameters. Notably, LongCat-2.0 is the first model in the industry to successfully complete the entire process from training to inference on a domestic computing cluster consisting of 50,000 graphics cards. In terms of technical features, the model natively supports an ultra-long context window of up to 1M tokens and is specifically optimized for Agentic Coding tasks. Currently, its preview version has ranked among the top globally in terms of call volume, demonstrating strong application potential and market attention.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 龙猫LongCatSource published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.