llm-benchmarks
githubdrose.io

gpt-oss-safeguard-20b Benchmarks


Measured on groq

17 runs
groq site
Avg Tokens / Second
67.30
Decode Tokens / Second
Avg Time to First Token (ms)
940.00
Runs Analysed
17
Last Updated
Aug 6, 2026, 01:57 PM

Throughput distribution

Throughput over time

gpt-oss-safeguard-20bgroq323tok/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 432 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
groqvia OpenRouter67.3042.00110.00940.0017
groqdirect86.1086.0086.20650.002

Frequently Asked Questions

How fast is gpt-oss-safeguard-20b?
The latest rolling average throughput is 67.30 tokens per second with an average time to first token of 940.00 ms across 17 recent runs.
How often are these benchmarks updated?
Benchmarks refresh automatically whenever the monitoring cron runs. The most recent run completed on Aug 6, 2026, 01:57 PM.

Related models