Article complete

Meituan Open-Sources LongCat-2.0 Model Trained Entirely on Chinese AI Chips

The model supports a one-million-token context window and introduces a new sparse attention architecture for coding and agentic workloads.

Colorful graphic featuring a stylized dragon's face with text overlays in various colors and arrangements.

Key takeaways

  • Meituan has open-sourced LongCat-2.0, a large language model with 1.6 trillion parameters, trained entirely on Chinese AI chips.
  • The model features a one-million-token context window and a new sparse attention architecture for improved performance in coding tasks.
  • LongCat-2.0 reportedly matches or exceeds the performance of leading proprietary models, although these claims are yet to be independently verified.
  • The release highlights China's shift towards self-reliance in AI technology, particularly in response to US export restrictions on advanced AI hardware.
  • LongCat-2.0 is available for developers on platforms like GitHub and Hugging Face, providing API access for broader use.

Takeaways drafted with AI from the article text.

Chinese technology company Meituan has open-sourced LongCat-2.0, a 1.6 trillion-parameter mixture-of-experts (MoE) large language model. The company says the model was trained and deployed entirely on domestically developed AI chips, underscoring the country's efforts to build AI infrastructure independent of NVIDIA hardware.

LongCat-2.0 is available to download on GitHub and Hugging Face, along with API access for developers.

According to Meituan, LongCat-2.0 matches or exceeds proprietary models such as GPT-5.5, Gemini 3.1 Pro and Claude Opus on several coding and agent benchmarks. These claims have not yet been independently verified.

“Both the full training run and the large-scale deployment are built entirely on AI ASIC superpods,” the company announced. Pretraining spans millions of accelerator-days across more than 35 trillion tokens, with no rollbacks or irrecoverable loss spikes. 

The release comes as Chinese AI companies increasingly shift toward homegrown chips following US export restrictions on advanced AI chips. While companies such as DeepSeek have previously used domestic hardware primarily for inference, Meituan claims LongCat-2.0 is the first trillion-parameter model to complete both pre-training and inference entirely on Chinese AI hardware.

It further said LongCat-2.0 activates around 48 billion parameters per token and was pretrained on more than 35 trillion tokens across millions of accelerator-days using AI ASIC superpods.

LongCat-2.0 supports a native one-million-token context window and introduces LongCat Sparse Attention (LSA), an updated sparse attention mechanism that combines streaming-aware, cross-layer and hierarchical indexing to reduce long-context inference costs. 

The model also incorporates an N-gram Embedding module, which expands the embedding space by nearly 100 times while accounting for less than 10% of the model's total parameters.

To support the model’s scale, Meituan said it trained LongCat-2.0 on more than 50,000 AI ASICs using a six-dimensional parallelism strategy, deterministic training operators, memory optimisations, and fault-recovery systems. The company said these optimisations improved training throughput by over 35% compared with a baseline implementation.

Meituan is positioning LongCat-2.0 as an agentic coding model. It said the model integrates with developer tools, including Claude Code, OpenClaw, and Hermes, for repository-level code editing, automated task execution, and software engineering workflows. The company also introduced three post-training expert groups focused on reasoning, agent execution and interaction quality.



Published

Corrections: AIM corrects errors of fact and notes the change on the story. Spotted one? Tell the editors. Read our editorial standards.

More in this story

AI · 5 stories
  1. You are here: Meituan Open-Sources LongCat-2.0 Model Trained Entirely on Chinese AI Chips
  2. Meta Expands AI Age-Estimation Tech to Protect Teens on Instagram
  3. IBM, Microsoft Launch Agentic AI Platform With Over 6,000 AI Agents
  4. Google Launches Gemini 4 Hydrogen for Faster Coding and AI Workflows
  5. Adobe, Google Expand AI Partnership Across Creative and Productivity Tools
About the author

Tech Journalist

Siddharth is a media graduate who loves to explore tech through journalism and putting forward ideas worth pondering about in the era of artificial intelligence.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.