GLM-5 Turbo from Z-AI is live on Chat-O. It’s a new model from Z-AI designed specifically for fast inference and strong performance in agent-driven environments, deeply optimized for scenarios like OpenClaw where the model has to make many quick decisions in a row.

Why a Turbo variant?

Reasoning models like GLM-5 and GLM-5.1 are thorough but slow. That’s fine for big one-shot questions but costly when an agent loop calls the model dozens of times per task.

GLM-5 Turbo keeps the agent-friendly behavior of the GLM-5 family but trades a little depth for a lot of speed, making it the right pick when:

• An agent workflow makes many small decisions per run
• You need high-volume inference without Power-tier pricing
• The task is clear enough that you don’t need maximum reasoning depth

Where it fits

GLM-5 Turbo is the only GLM model in the Light tier, making it the most economical way to run the GLM-5 family on Chat-O.

Try it

Pick GLM-5 Turbo in the model dropdown. Light-tier credits apply — perfect for high-volume agent loops and fast chat.