llm-benchmarks
githubdrose.io

deepseek Provider Benchmarks


Measured over the last 30 days

Models Tracked
1
Avg Tokens / Second
6.92
Avg Time to First Token (ms)
8930.00
Last Updated
Oct 14, 2026

All deepseek models

1 live · one row per model · fastest first
deepseek site
ModelAvg Toks/SecMinMaxAvg TTF (ms)
DeepSeek V4.1 Flash6.926.926.928930.00

Frequently Asked Questions

Which deepseek model is fastest?
Based on recent tests, DeepSeek V4.1 Flash shows the highest average throughput among tracked deepseek models.
How many recent measurements feed this dashboard?
This provider summary aggregates 801 individual prompts measured across 776 monitoring runs over the past month.