Celeris-1 vs Granite 4.2 30B
Which is better for the money: Celeris-1 by Celeris or Granite 4.2 30B by IBM? 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
Granite 4.2 30B scores 6.4 points higher (12.7 vs 6.3) but costs 1.8× as much to run the benchmark ($111 vs $60.0; both mid-range class). Pick Granite 4.2 30B when the extra intelligence is worth the premium, otherwise Celeris-1.
| Metric | Celeris-1 | Granite 4.2 30B | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 12.7 | +101.6% Granite 4.2 30B is 2.0× more intelligent |
| Total AA cost (lower is better) | $59.99 | $110.83 | +84.7% Granite 4.2 30B is 84.7% more expensive |
| End-to-end time (lower is better) | 0.87s | 44.22s | +4982.8% Granite 4.2 30B is 4982.8% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $0.16 / $0.65 | |
| Input modalities | Text, Image | Text |
Granite 4.2 30B vs Celeris-1: 101.6% more intelligence, 84.7% more cost to run the AA benchmark, and 4982.8% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Celeris-1 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Medium) scores 34.50 for $56.4.
- Granite 4.2 30B is off the frontier: at the same or lower cost, MiMo-V2.6-Flash scores 37.90 for $110.
Celeris-1Granite 4.2 30BToday'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/celeris-1/granite-4-2-30b · Markdown: /api/v1/compare/celeris-1/granite-4-2-30b?format=md · llms.txt · agent skill