ZAI Monitor

Coding Plan performance

Directional benchmarking of Z.AI inference behavior across models and rolling time windows.

Current Snapshot

Latest values
Tokens per Second Latest

GLM-5

—

GLM-4.7

—

GLM-4.7-Flash

—

Time to First Token Latest

GLM-5

—

GLM-4.7

—

GLM-4.7-Flash

—

Historical Window

Last 24h

Historical Trends

No trend data in this window for Tokens/sec.

Success Rate

GLM-5

—

GLM-4.7

—

GLM-4.7-Flash

—

p95 Time to First Token

GLM-5

—

GLM-4.7

—

GLM-4.7-Flash

—

End-to-End Throughput Avg

GLM-5

—

GLM-4.7

—

GLM-4.7-Flash

—

completion tokens / total latency

Methodology