llm-benchmarks
githubdrose.io

Gemma 4 26B A4B Benchmarks


Measured on dekallm

1 runs
Avg Tokens / Second
43.30
Decode Tokens / Second
Avg Time to First Token (ms)
840.00
Runs Analysed
1
Last Updated
Sep 11, 2026, 04:34 PM

Throughput distribution

Throughput over time

Gemma 4 26B A4Bdekallm43tok/s

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 45 tok/s (99th percentile) · dashed rule is the model's own mean · 1 cell carry fewer than 4 samples

Every lane measuring this model

1 measured · live first · a lane is a provider, a catalogue id and a transport
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
dekallmvia OpenRouter43.3043.3043.30840.001

Frequently Asked Questions

How fast is Gemma 4 26B A4B?
The latest rolling average throughput is 43.30 tokens per second with an average time to first token of 840.00 ms across 1 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Sep 11, 2026, 04:34 PM.

Related models