@sheriyuo: The industry's first trillion-parameter model to complete end-to-end training and inference on a 50,000-GPU Chinese com…

X AI KOLs Timeline Models

Summary

Meituan released LongCat-2.0, a 1.6T-parameter MoE model with 1M context, claimed as the first to train on a 50,000-GPU Chinese cluster, now available on OpenRouter for agentic coding.

The industry's first trillion-parameter model to complete end-to-end training and inference on a 50,000-GPU Chinese computing cluster 👀
Original Article
View Cached Full Text

Cached at: 06/30/26, 05:36 AM

The industry’s first trillion-parameter model to complete end-to-end training and inference on a 50,000-GPU Chinese computing cluster 👀

Meituan LongCat (@Meituan_LongCat): Introducing LongCat-2.0 🐱 1.6T parameters · MoE with ~48B active · 1M context The full model behind Owl Alpha on @OpenRouter — now available.

Built for agentic coding from the ground up: ◆ LongCat Sparse Attention (LSA) — scales efficiently for 1M-context tokens ◆

Similar Articles