MiMo-V2.5 vs Ministral 3 14B
Which is better for the money: MiMo-V2.5 by Xiaomi or Ministral 3 14B 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
MiMo-V2.5 scores 15.8 points higher (21.8 vs 6.0) but costs 1.6× as much to run the benchmark ($61.9 vs $37.6; mid-range vs budget cost classes). Pick MiMo-V2.5 when the extra intelligence is worth the premium, otherwise Ministral 3 14B.
| Metric | MiMo-V2.5 | Ministral 3 14B | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 21.8 | 6.0 | −72.5% Ministral 3 14B is 72.5% less intelligent |
| Total AA cost (lower is better) | $61.93 | $37.56 | −39.4% Ministral 3 14B is 39.4% cheaper |
| End-to-end time (lower is better) | 66.80s | 8.05s | −87.9% Ministral 3 14B is 87.9% faster |
| Price per 1M tokens (in / out) | $0.13 / $0.27 | $0.20 / $0.20 | |
| Input modalities | Text, Image | Text, Image |
Ministral 3 14B vs MiMo-V2.5: 72.5% less intelligence, 39.4% less cost to run the AA benchmark, and 87.9% faster end-to-end.
Today's Pareto frontier
Against the frontier
- MiMo-V2.5 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Medium) scores 34.50 for $56.4.
- Ministral 3 14B is off the frontier: at the same or lower cost, GPT-6 Luna (Medium) scores 29.90 for $31.2.
MiMo-V2.5Ministral 3 14BToday'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/mimo-v2-5-0424/ministral-3-14b · Markdown: /api/v1/compare/mimo-v2-5-0424/ministral-3-14b?format=md · llms.txt · agent skill