llm-benchmarks
githubdrose.io

claude-opus-4.8 Benchmarks


Measured on claudeonaws

1 runs
Avg Tokens / Second
25.80
Decode Tokens / Second
—
Avg Time to First Token (ms)
1290.00
Runs Analysed
1
Last Updated
Sep 22, 2026, 05:57 AM

Throughput distribution

Throughput over time

claude-opus-4.8claudeonaws26tok/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 27 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
claudeonawsvia OpenRouter25.8025.8025.801290.001

Frequently Asked Questions

How fast is claude-opus-4.8?
The latest rolling average throughput is 25.80 tokens per second with an average time to first token of 1290.00 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 22, 2026, 05:57 AM.

Related models