Celeris-1 vs Nemotron 3.5 Super VL
Which is better for the money: Celeris-1 by Celeris or Nemotron 3.5 Super VL by NVIDIA? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Trade-off
Verdict
Depends on your budget
Nemotron 3.5 Super VL scores 20.2 points higher (26.5 vs 6.3) but costs 4.8× as much to run the benchmark ($286 vs $60.0; both mid-range class). They are not alternatives at one budget: pick Nemotron 3.5 Super VL when you need intelligence above 6.3, and Celeris-1 when 6.3 is enough.
| Metric | Celeris-1 | Nemotron 3.5 Super VL | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 26.5 | +320.6% Nemotron 3.5 Super VL is 4.2× more intelligent |
| Total AA cost (lower is better) | $59.99 | $286.07 | +376.9% Nemotron 3.5 Super VL is 4.8× more expensive |
| End-to-end time (lower is better) | 0.86s | 6.32s | +634.9% Nemotron 3.5 Super VL is 634.9% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $0.23 / $0.57 | |
| Input modalities | Text, Image | Text, Image |
Nemotron 3.5 Super VL vs Celeris-1: 320.6% more intelligence, 376.9% more cost to run the AA benchmark, and 634.9% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Celeris-1 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Medium) scores 34.50 for $56.4.
- Nemotron 3.5 Super VL is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
Celeris-1Nemotron 3.5 Super VLToday's Pareto frontierOpen-weights frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/celeris-1/nemotron-3-5-super-vl · Markdown: /api/v1/compare/celeris-1/nemotron-3-5-super-vl?format=md · llms.txt · agent skill