Benchmark Overview
21.90
770.00
208
Apr 13, 2026, 06:03 AM
Key Insights
gpt-4 streams at 21.90 tokens/second on average across the last 208 benchmark runs.
Performance fluctuated by 40.41 tokens/second (184.5% coefficient of variation), indicating variable behavior across benchmark runs.
Average time to first token is 770.00 ms (good latency), suitable for latency-sensitive workloads.
Latest measurements completed on Apr 13, 2026, 06:03 AM based on 208 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.
gpt-4
Benchmark Samples
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| openai | gpt-4 | 21.90 | 4.09 | 44.50 | 770.00 |
Frequently Asked Questions
The latest rolling average throughput is 21.90 tokens per second with an average time to first token of 770.00 ms across 208 recent runs.
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Apr 13, 2026, 06:03 AM.