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, gemma-4-31b-it leads 0.
| laguna-xs-2.1 | Metric | gemma-4-31b-it |
|---|---|---|
| poolside Serving | Status | Unstable google |
| 90● | Score | 31 |
| 100%● | Uptime | 33.3% |
| 171 ms● | First token (median) | 59.1 s |
| 4.91 s● | Full reply (median) | 66.2 s |
| 2.07 s● | Best reply | 40.6 s |
| 31.0 tok/s● | Throughput (median) | 20.5 tok/s |
| 0 | Round wins | 0 |
| #5● | Leaderboard rank | #11 |
| 100% | Faster when both answered (4 rounds) | 0% |
marks the better value in each row
Full reply time overlay
Each probe, both models on one axis. Lower is better.
Win timeline
Per round, oldest left: which model replied faster.
AAAAAAAAAAAA
Alaguna-xs-2.1 Bgemma-4-31b-it ·neither answered