Ministral 3 8B vs Mistral Large 4 Preview
Which is better for the money: Ministral 3 8B by Mistral or Mistral Large 4 Preview 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
Mistral Large 4 Preview scores 32.9 points higher (38.4 vs 5.5) but costs 59× as much to run the benchmark ($1608 vs $27.5; premium vs budget cost classes). They are not alternatives at one budget: pick Mistral Large 4 Preview when you need intelligence above 5.5, and Ministral 3 8B when 5.5 is enough.
| Metric | Ministral 3 8B | Mistral Large 4 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 5.5 | 38.4 | +598.2% Mistral Large 4 Previ… is 7.0× more intelligent |
| Total AA cost (lower is better) | $27.47 | $1,607.86 | +5753.1% Mistral Large 4 Previ… is 59× more expensive |
| End-to-end time (lower is better) | 4.92s | 25.22s | +412.6% Mistral Large 4 Previ… is 412.6% slower |
| Price per 1M tokens (in / out) | $0.15 / $0.15 | $1.36 / $4.18 | |
| Input modalities | Text, Image | Text, Image |
Mistral Large 4 Preview vs Ministral 3 8B: 598.2% more intelligence, 5753.1% more cost to run the AA benchmark, and 412.6% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Ministral 3 8B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Mistral Large 4 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Ministral 3 8BMistral Large 4 PreviewToday'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-3-8b/mistral-large-4 · Markdown: /api/v1/compare/ministral-3-8b/mistral-large-4?format=md · llms.txt · agent skill