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 7 days.
Head-to-head stats
nemotron-3.5-bolt-e4b leads 4 rows, nemotron-3.5-super-vl-120b-a12b leads 3.
| nemotron-3.5-bolt-e4b | Metric | nemotron-3.5-super-vl-120b-a12b |
|---|---|---|
| nvidia Serving | Status | Serving nvidia |
| 95● | Score | 93 |
| 100% | Uptime | 100% |
| 114 ms● | First token (median) | 343 ms |
| 1.63 s● | Full reply (median) | 2.00 s |
| 1.46 s | Best reply | ●957 ms |
| 86.3 tok/s | Throughput (median) | ●126 tok/s |
| 1 | Round wins | ●2 |
| #1● | Leaderboard rank | #2 |
| 73.3% | Faster when both answered (15 rounds) | 26.7% |
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.
AAABAABBAABAAAA
Anemotron-3.5-bolt-e4b Bnemotron-3.5-super-vl-120b-a12b ·neither answered