llm-benchmarks
githubdrose.io

mara Provider Benchmarks


Measured over the last 30 days

Models Tracked
3
Avg Tokens / Second
22.40
Avg Time to First Token (ms)
6158.81
Last Updated
Sep 1, 2026

All mara models

3 tracked · fastest first
ModelAvg Toks/SecMinMaxAvg TTF (ms)
gpt-oss-120b87.1073.8099.20760.00
DeepSeek V3.133.704.5649.504300.00
MiniMax M2.729.305.8041.104450.00

Frequently Asked Questions

Which mara model is fastest?
Based on recent tests, gpt-oss-120b shows the highest average throughput among tracked mara models.
How many recent measurements feed this dashboard?
This provider summary aggregates 410 individual prompts measured across 373 monitoring runs over the past month.