Benchmark Overview
28.70
650.00
209
Apr 26, 2026, 03:03 PM
Key Insights
GPT-5.3-codex streams at 28.70 tokens/second on average across the last 209 benchmark runs.
Performance fluctuated by 34.30 tokens/second (119.5% coefficient of variation), indicating variable behavior across benchmark runs.
Average time to first token is 650.00 ms (good latency), suitable for latency-sensitive workloads.
Latest measurements completed on Apr 26, 2026, 03:03 PM based on 209 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-5.3-codex
Benchmark Samples
| Provider | Model | Avg Toks/Sec | Min | Max | Avg TTF (ms) |
|---|---|---|---|---|---|
| openai | GPT-5.3-codex | 28.70 | 8.10 | 42.40 | 650.00 |
Frequently Asked Questions
The latest rolling average throughput is 28.70 tokens per second with an average time to first token of 650.00 ms across 209 recent runs.
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Apr 26, 2026, 03:03 PM.