MiniMax-M3 vs Mistral Large 4 Preview
Which is better for the money: MiniMax-M3 by MiniMax or Mistral Large 4 Preview by Mistral? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-29.
Trade-off
Verdict
Depends on your budget
Mistral Large 4 Preview scores 6.6 points higher (25.3 vs 18.7) but costs 3.3× as much to run the benchmark ($2923 vs $878; flagship vs premium cost classes). They are not alternatives at one budget: pick Mistral Large 4 Preview when you need intelligence above 18.7, and MiniMax-M3 when 18.7 is enough.
| Metric | MiniMax-M3 | Mistral Large 4 Preview | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 18.7 | 25.3 | +35.3% Mistral Large 4 Previ… is 35.3% more intelligent |
| Total AA cost (lower is better) | $878.03 | $2,923.39 | +232.9% Mistral Large 4 Previ… is 3.3× more expensive |
| End-to-end time (lower is better) | 25.42s | 25.54s | +0.5% Mistral Large 4 Previ… is 0.5% slower |
| Price per 1M tokens (in / out) | $0.30 / $1.20 | $0.70 / $2.10 | |
| Input modalities | Text, Image, Video | Text, Image |
Mistral Large 4 Preview vs MiniMax-M3: 35.3% more intelligence, 232.9% more cost to run the AA benchmark, and 0.5% slower end-to-end.
Today's Pareto frontier
Against the frontier
- MiniMax-M3 is off the frontier: at the same or lower cost, GPT-6.1 Sol (Xhigh) scores 52.30 for $718.
- Mistral Large 4 Preview is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 54.50 for $1124.
MiniMax-M3Mistral 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/minimax-m3/mistral-large-4-preview · Markdown: /api/v1/compare/minimax-m3/mistral-large-4-preview?format=md · llms.txt · agent skill