Celeris-1 vs Ministral 3 3B
Which is better for the money: Celeris-1 by Celeris or Ministral 3 3B by Mistral? 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
Celeris-1 scores 1.5 points higher (6.3 vs 4.8) but costs 4.9× as much to run the benchmark ($60.0 vs $12.2; mid-range vs budget cost classes). They are not alternatives at one budget: pick Celeris-1 when you need intelligence above 4.8, and Ministral 3 3B when 4.8 is enough.
| Metric | Celeris-1 | Ministral 3 3B | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 4.8 | −23.8% Ministral 3 3B is 23.8% less intelligent |
| Total AA cost (lower is better) | $59.99 | $12.23 | −79.6% Ministral 3 3B is 79.6% cheaper |
| End-to-end time (lower is better) | 0.87s | 3.08s | +254.0% Ministral 3 3B is 254.0% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $0.10 / $0.10 | |
| Input modalities | Text, Image | Text, Image |
Ministral 3 3B vs Celeris-1: 23.8% less intelligence, 79.6% less cost to run the AA benchmark, and 254.0% 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.
- Ministral 3 3B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
Celeris-1Ministral 3 3BToday'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/ministral-3-3b · Markdown: /api/v1/compare/celeris-1/ministral-3-3b?format=md · llms.txt · agent skill