Granite 4.2 3B vs Mercury 2
Which is better for the money: Granite 4.2 3B by IBM or Mercury 2 by Inception? 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
Mercury 2 scores 1.7 points higher (10.8 vs 9.1) but costs 7.0× as much to run the benchmark ($105 vs $14.9; mid-range vs budget cost classes). They are not alternatives at one budget: pick Mercury 2 when you need intelligence above 9.1, and Granite 4.2 3B when 9.1 is enough.
| Metric | Granite 4.2 3B | Mercury 2 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 9.1 | 10.8 | +18.7% Mercury 2 is 18.7% more intelligent |
| Total AA cost (lower is better) | $14.91 | $104.94 | +603.8% Mercury 2 is 7.0× more expensive |
| End-to-end time (lower is better) | 49.98s | 6.56s | −86.9% Mercury 2 is 86.9% faster |
| Price per 1M tokens (in / out) | $0.03 / $0.12 | $0.25 / $0.75 | |
| Input modalities | Text | Text |
Mercury 2 vs Granite 4.2 3B: 18.7% more intelligence, 603.8% more cost to run the AA benchmark, and 86.9% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Granite 4.2 3B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Mercury 2 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
Granite 4.2 3BMercury 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/granite-4-2-3b/mercury-2 · Markdown: /api/v1/compare/granite-4-2-3b/mercury-2?format=md · llms.txt · agent skill