Granite 4.2 30B vs Ministral 3 8B
Which is better for the money: Granite 4.2 30B by IBM or Ministral 3 8B 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
Granite 4.2 30B scores 7.2 points higher (12.7 vs 5.5) but costs 4.0× as much to run the benchmark ($111 vs $27.5; mid-range vs budget cost classes). They are not alternatives at one budget: pick Granite 4.2 30B when you need intelligence above 5.5, and Ministral 3 8B when 5.5 is enough.
| Metric | Granite 4.2 30B | Ministral 3 8B | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 12.7 | 5.5 | −56.7% Ministral 3 8B is 56.7% less intelligent |
| Total AA cost (lower is better) | $110.83 | $27.47 | −75.2% Ministral 3 8B is 75.2% cheaper |
| End-to-end time (lower is better) | 65.19s | 5.26s | −91.9% Ministral 3 8B is 91.9% faster |
| Price per 1M tokens (in / out) | $0.16 / $0.65 | $0.15 / $0.15 | |
| Input modalities | Text | Text, Image |
Ministral 3 8B vs Granite 4.2 30B: 56.7% less intelligence, 75.2% less cost to run the AA benchmark, and 91.9% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Granite 4.2 30B is off the frontier: at the same or lower cost, MiMo-V2.6-Flash scores 37.90 for $110.
- Ministral 3 8B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
Granite 4.2 30BMinistral 3 8BToday'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/granite-4-2-30b/ministral-3-8b · Markdown: /api/v1/compare/granite-4-2-30b/ministral-3-8b?format=md · llms.txt · agent skill