Official Z.ai platform to experience our new, MIT-licensed GLM models (Base, Reasoning, Rumination). Simple UI focuses on model interaction. Free.
This is the 11th launch from Z.ai. View more
GLM-5.1
Launching today
GLM-5.1 is a next-gen AI model built for agentic engineering, excelling at long-horizon coding tasks. It iterates, tests, and optimizes over hundreds of steps, outperforming prior models on real-world benchmarks and complex software workflows.
Free Options
Launch Team
GLM-5.1 is a next-gen AI model built for agentic engineering and long-horizon coding tasks, tackling the key limitation of current models that plateau after initial progress. Instead, it keeps improving over hundreds of iterations with smarter reasoning and self-correction.
What stands out is its ability to sustain performance over long sessions, breaking down complex problems, running experiments, and iterating toward better results—something most models struggle with.
Key features & benefits:
Strong coding + repo generation capabilities
Excels in real-world terminal & software engineering tasks
Long-horizon optimization (thousands of tool calls)
Open-source (MIT) + flexible deployment (local, API, agents)
Perfect for developers, AI engineers, and teams working on complex systems, automation, and advanced coding workflows.
@GitHub: https://github.com/zai-org/GLM-5
@Hugging Face: https://huggingface.co/zai-org/GLM-5.1
If you’re building with AI agents or pushing coding automation forward, this is definitely worth exploring.