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

gemma-4-31b-it leads 1 rows, muse-glimmer-30b leads 6.

gemma-4-31b-itMetricmuse-glimmer-30b
google UnstableStatusServing meta
31Score●64
33.3%Uptime●91.7%
59.1 sFirst token (median)●33.4 s
66.2 sFull reply (median)●61.1 s
40.6 sBest reply●26.5 s
20.5 tok/s●Throughput (median)8.2 tok/s
0Round wins0
#11Leaderboard rank●#9
50.0%Faster when both answered (4 rounds)50.0%

marks the better value in each row

Full reply time overlay

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

gemma-4-31b-itmuse-glimmer-30b 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.

Agemma-4-31b-it Bmuse-glimmer-30b ·neither answered