llm-benchmarks
githubdrose.io

z-ai Provider Benchmarks


Measured over the last 30 days

Models Tracked
11
Avg Tokens / Second
12.70
Avg Time to First Token (ms)
18284.62
Last Updated
Sep 1, 2026

All z-ai models

11 tracked · fastest first
ModelAvg Toks/SecMinMaxAvg TTF (ms)
GLM 5.212.309.6816.103900.00
GLM 4.5 Air9.259.029.455750.00
GLM 5 Turbo9.061.5514.7012500.00
GLM 4.5V3.782.505.6018000.00
GLM 5V Turbo3.172.124.7721600.00
GLM 4.52.642.033.1825600.00
GLM 52.531.933.7328500.00
GLM 4.72.321.183.3332500.00
GLM 4.62.161.083.4936400.00
GLM 4.6V1.461.461.4641800.00
GLM 5.11.251.151.3449400.00

Frequently Asked Questions

Which z-ai model is fastest?
Based on recent tests, GLM 5.2 shows the highest average throughput among tracked z-ai models.
How many recent measurements feed this dashboard?
This provider summary aggregates 2563 individual prompts measured across 2472 monitoring runs over the past month.