llm-benchmarks
githubdrose.io

anthropic Provider Benchmarks


Measured over the last 30 days

Models Tracked
10
Avg Tokens / Second
25.28
Avg Time to First Token (ms)
1745.00
Last Updated
Aug 9, 2026

All anthropic models

10 tracked · fastest first
anthropic site
ModelAvg Toks/SecMinMaxAvg TTF (ms)
claude-haiku-4.543.406.2666.30890.00
claude-sonnet-528.406.5342.801150.00
claude-opus-4-827.603.8338.701460.00
claude-opus-525.605.2735.702620.00
claude-opus-4-724.604.9534.201470.00
claude-opus-4.524.505.2836.001270.00
claude-sonnet-4.623.905.5132.301240.00
claude-fable-520.707.3328.803260.00
claude-opus-4.618.706.4024.101930.00
Claude Opus 4.115.407.2119.102160.00

Frequently Asked Questions

Which anthropic model is fastest?
Based on recent tests, claude-haiku-4.5 shows the highest average throughput among tracked anthropic models.
How many recent measurements feed this dashboard?
This provider summary aggregates 2705 individual prompts measured across 2304 monitoring runs over the past month.