llm-benchmarks
githubdrose.io

Lyria 3 Clip Preview Benchmarks


Measured on google

4 runs
google site
Avg Tokens / Second
34.40
Decode Tokens / Second
Avg Time to First Token (ms)
3750.00
Runs Analysed
4
Last Updated
Sep 1, 2026, 03:38 AM

Throughput distribution

Throughput over time

Lyria 3 Clip Previewgoogle35tok/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 38 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
ProviderLaneAvg Toks/SecMinMaxAvg TTF (ms)n
googlevia OpenRouter34.4029.7036.303750.004
googledirect37.8025.2062.803810.0037

Frequently Asked Questions

How fast is Lyria 3 Clip Preview?
The latest rolling average throughput is 34.40 tokens per second with an average time to first token of 3750.00 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 Sep 1, 2026, 03:38 AM.

Related models