Zhipu AI Releases Open-Weight MoE Model GLM-4.6
The new Mixture-of-Experts model is available under a permissive MIT license and is optimized for complex reasoning and coding tasks.

Zhipu AI, one of China's leading AI labs, has released GLM-4.6, a new large language model with open weights. The model utilizes a Mixture-of-Experts (MoE) architecture, a design known for balancing high performance with more efficient inference compared to dense models of similar capability.
GLM-4.6 is positioned as a strong performer in complex tasks that require sophisticated logical deduction. According to the development team, the model shows significant improvements in both logical reasoning and code generation abilities, making it a powerful tool for developers and researchers.
Why it matters
Perhaps most notable is the model's release under the MIT license. This permissive license allows for broad use, including commercial applications, removing a common barrier for businesses looking to build on top of state-of-the-art open models. The release of a capable MoE model from a major international lab under such terms is a significant contribution to the open-source ecosystem.
Researchers and developers can access the model weights and further details directly from the official repository on Hugging Face. The release provides another powerful, commercially-friendly foundation model for the global AI community.
Sources
- Visit
zai-org/GLM-4.6
Hugging Face
More in Text / LLM
Meituan Ships a Lighter, Sparser LongCat-Flash
The food-delivery giant's newest open model trims its mixture-of-experts design for more efficient inference under an MIT license.
DeepSeek Refreshes V4-Flash With New 0731 Checkpoint
The MIT-licensed mixture-of-experts model returns in an updated build shipping with FP8 weights for cheaper inference.
DeepSeek Ships V4-Flash, a 304B MoE Tuned for Agents
The latest checkpoint in DeepSeek's V4 line leans into agentic workflows while keeping the permissive MIT license.
0 comments
No comments yet. Be the first to weigh in.