GPT-5.6 Terra (Xhigh) vs Granite 4.2 3B
Which is better for the money: GPT-5.6 Terra (Xhigh) by OpenAI or Granite 4.2 3B by IBM? 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
GPT-5.6 Terra (Xhigh) scores 28.9 points higher (38.0 vs 9.1) but costs 80× as much to run the benchmark ($1187 vs $14.9; premium vs budget cost classes). They are not alternatives at one budget: pick GPT-5.6 Terra (Xhigh) when you need intelligence above 9.1, and Granite 4.2 3B when 9.1 is enough.
| Metric | GPT-5.6 Terra (Xhigh) | Granite 4.2 3B | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 38.0 | 9.1 | −76.1% Granite 4.2 3B is 76.1% less intelligent |
| Total AA cost (lower is better) | $1,186.77 | $14.91 | −98.7% Granite 4.2 3B is 98.7% cheaper |
| End-to-end time (lower is better) | 36.37s | 63.70s | +75.1% Granite 4.2 3B is 75.1% slower |
| Price per 1M tokens (in / out) | $2.00 / $12.0 | $0.03 / $0.12 | |
| Input modalities | Text, Image | Text |
Granite 4.2 3B vs GPT-5.6 Terra (Xhigh): 76.1% less intelligence, 98.7% less cost to run the AA benchmark, and 75.1% slower end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-5.6 Terra (Xhigh) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- Granite 4.2 3B is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
GPT-5.6 Terra (Xhigh)Granite 4.2 3BToday'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/gpt-5-6-terra-xhigh/granite-4-2-3b · Markdown: /api/v1/compare/gpt-5-6-terra-xhigh/granite-4-2-3b?format=md · llms.txt · agent skill