Z.ai: GLM 5

GLM 5 from Z.ai is a text-in, text-out model with a large 202,752-token context window. It supports tool use and reasoning, making it usable for multi-step agentic workflows and tasks that benefit from extended document context. Structured output support is unconfirmed, so developers who depend on strict JSON schemas should verify compatibility before committing. On the comparison side, GLM 5 sits at $0.60 per million input tokens and $1.92 per million output tokens, which is competitive for a reasoning-capable model. Its blended benchmark score of 76.2 comes from only two benchmarks, so treat that figure as a preliminary signal rather than a settled verdict. Teams doing long-context summarization or tool-augmented pipelines on a moderate budget have reason to shortlist it, but those needing broad, validated performance evidence may want to wait for wider benchmark coverage.