- Avg Tokens / Second
- 2.90
- Decode Tokens / Second
- —
- Avg Time to First Token (ms)
- 46200.00
- Runs Analysed
- 8
- Last Updated
- Sep 21, 2026, 04:15 PM
Throughput distribution
Throughput over time
Generated throughput, including reasoning tokens — so a thinking model's line does not collapse; the table ranks on visible tokens · one cell per model, showing its best-covered provider · models with at least 4 samples in the window first, fastest of those at the top · shared vertical scale, 0 to 47 tok/s (99th percentile) · dashed rule is the model's own mean
Every lane measuring this model
2 measured · live first · a lane is a provider, a catalogue id and a transport| Provider | Lane | Avg Toks/Sec | Min | Max | Avg TTF (ms) | n |
|---|---|---|---|---|---|---|
| deepinfra | via OpenRouter | 2.90 | 0.81 | 11.90 | 46200.00 | 8 |
| deepinfra | direct | 4.71 | 0.98 | 8.44 | 34700.00 | 2 |
Frequently Asked Questions
How fast is Qwen3.8 27B?
The latest rolling average throughput is 2.90 tokens per second with an average time to first token of 46200.00 ms across 8 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Sep 21, 2026, 04:15 PM.