Ministral 3 3B vs Mistral Small 3.2
Which is better for the money: Ministral 3 3B by Mistral or Mistral Small 3.2 by Mistral? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Trade-off
Verdict
Depends on your budget
Mistral Small 3.2 scores 2.2 points higher (7.0 vs 4.8) but costs 10.0× as much to run the benchmark ($122 vs $12.2; mid-range vs budget cost classes). They are not alternatives at one budget: pick Mistral Small 3.2 when you need intelligence above 4.8, and Ministral 3 3B when 4.8 is enough.
| Metric | Ministral 3 3B | Mistral Small 3.2 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 4.8 | 7.0 | +45.8% Mistral Small 3.2 is 45.8% more intelligent |
| Total AA cost (lower is better) | $12.23 | $121.78 | +895.7% Mistral Small 3.2 is 10.0× more expensive |
| End-to-end time (lower is better) | 3.06s | 9.55s | +212.1% Mistral Small 3.2 is 212.1% slower |
| Price per 1M tokens (in / out) | $0.10 / $0.10 | $0.07 / $0.20 | |
| Input modalities | Text, Image | Text, Image |
Mistral Small 3.2 vs Ministral 3 3B: 45.8% more intelligence, 895.7% more cost to run the AA benchmark, and 212.1% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Ministral 3 3B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Mistral Small 3.2 is off the frontier: at the same or lower cost, MiMo-V2.6-Flash scores 37.90 for $110.
Ministral 3 3BMistral Small 3.2Today'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/ministral-3-3b/mistral-small-3-2 · Markdown: /api/v1/compare/ministral-3-3b/mistral-small-3-2?format=md · llms.txt · agent skill