Skip to main content

Includes 694 hourly rounds up to 2026-11-16 from NIMStats by Mauro Druwel (source), measured with a different prompt and timeout. Our own probe adds a round every 30 minutes.

Model comparison

Head-to-head from live probes, last 24 hours.

⇄

Head-to-head stats

laguna-xs-2.1 leads 7 rows, step-5-preview leads 0.

laguna-xs-2.1Metricstep-5-preview
poolside ServingStatusUnstable stepfun
90●Score45
100%●Uptime50.0%
171 ms●First token (median)7.49 s
4.91 s●Full reply (median)25.4 s
2.07 s●Best reply7.40 s
31.0 tok/s●Throughput (median)10.7 tok/s
0Round wins0
#5●Leaderboard rank#10
100%Faster when both answered (6 rounds)0%

marks the better value in each row

Full reply time overlay

Each probe, both models on one axis. Lower is better.

laguna-xs-2.1step-5-preview no answer
0 ms50.0 s100.0 s150.0 s200.0 s11-17 10:4311-17 16:13 UTC

Win timeline

Per round, oldest left: which model replied faster.

Alaguna-xs-2.1 Bstep-5-preview ·neither answered