Vercel

Zai Releases GLM 5.2 Model with 1M Token Context on AI Gateway


Executive Summary

Zai has launched GLM 5.2, a new large language model now available on the AI Gateway platform. Designed specifically for complex, long-horizon engineering tasks, the model features a significantly expanded 1M token context window. It aims to provide developers with more reliable execution for long-running processes and more consistent adherence to engineering standards.

Key Takeaways

* New Model: GLM 5.2 is officially available.

* Specialization: The model is optimized for project-level engineering tasks requiring deep, sustained context.

* Expanded Context Window: The context window has been increased five-fold to 1 million tokens, up from 200K in the previous version.

* Key Capabilities: It is built to run long tasks more reliably and follow engineering standards more consistently.

* Availability: GLM 5.2 is accessible immediately on AI Gateway and can be called via the AI SDK using the model identifier `zai/glm-5.2`.

* Platform Pricing: The AI Gateway platform charges no markup on provider pricing and no platform fees for inference.

Strategic Importance

This release positions AI Gateway as a platform for sophisticated, professional-grade AI development by offering a specialized model with a massive context window tailored for complex engineering and coding tasks.

Original article