Measured on venice
4 runs- Avg Tokens / Second
- 23.50
- Decode Tokens / Second
- —
- Avg Time to First Token (ms)
- 610.00
- Runs Analysed
- 4
- Last Updated
- Oct 12, 2026, 07:23 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 26 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 |
|---|---|---|---|---|---|---|
| venice | direct | 23.50 | 21.80 | 24.40 | 610.00 | 4 |
| venice | via OpenRouter | 22.50 | 18.20 | 32.40 | 1090.00 | 6 |
Frequently Asked Questions
How fast is Gemma 4 31B?
The latest rolling average throughput is 23.50 tokens per second with an average time to first token of 610.00 ms across 4 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Oct 12, 2026, 07:23 PM.