llm-benchmarks
githubdrose.io

decart Provider Benchmarks


Measured over the last 30 days

Models Tracked
6
Avg Tokens / Second
13.05
Avg Time to First Token (ms)
13418.14
Last Updated
Sep 1, 2026

All decart models

6 tracked · fastest first
ModelAvg Toks/SecMinMaxAvg TTF (ms)
GLM 5.224.9015.3034.902500.00
GLM 5.25.355.355.3512400.00
Kimi K2.63.521.416.3920600.00
DeepSeek V4 Flash 07313.513.513.5117900.00
DeepSeek V4 Flash 07313.441.679.0023000.00
Kimi K2.62.562.432.6923600.00

Frequently Asked Questions

Which decart model is fastest?
Based on recent tests, GLM 5.2 shows the highest average throughput among tracked decart models.
How many recent measurements feed this dashboard?
This provider summary aggregates 773 individual prompts measured across 633 monitoring runs over the past month.