Benchmark Overview
30.70
0.49
915
Oct 17, 2025, 08:07 AM
Key Insights
claude-3-5-haiku streams at 30.70 tokens/second on average across the last 915 benchmark runs.
Performance fluctuated by 32.73 tokens/second (106.6% coefficient of variation), indicating variable behavior across benchmark runs.
Average time to first token is 0.49 ms (excellent latency), suitable for latency-sensitive workloads.
Latest measurements completed on Oct 17, 2025, 08:07 AM based on 915 total samples.
Performance Distribution
Distribution of throughput measurements showing performance consistency across benchmark runs.
Performance Over Time
Historical performance trends showing how throughput has changed over the benchmarking period.
claude-3-5-haiku
Benchmark Samples
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| anthropic | claude-3-5-haiku | 30.70 | 8.67 | 41.40 | 0.49 |
Frequently Asked Questions
The latest rolling average throughput is 30.70 tokens per second with an average time to first token of 0.49 ms across 915 recent runs.
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Oct 17, 2025, 08:07 AM.