Skip to main content

Includes 693 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

step-5-preview leads 7 rows, gpt-oss-20b leads 0.

step-5-previewMetricgpt-oss-20b
stepfun UnstableStatusDown openai
43●Score0
46.2%●Uptime0%
7.49 s●First token (median)n/a
25.4 s●Full reply (median)n/a
7.40 s●Best replyn/a
10.7 tok/s●Throughput (median)n/a
0Round wins0
#10●Leaderboard rank#14
n/aFaster when both answered (0 rounds)n/a

marks the better value in each row

Full reply time overlay

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

step-5-previewgpt-oss-20b no answer
0 ms50.0 s100.0 s150.0 s200.0 s11-17 10:4311-17 16:43 UTC

Win timeline

Per round, oldest left: which model replied faster.

Astep-5-preview Bgpt-oss-20b ·neither answered