GPT-5.4 (Non-reasoning) vs Granite 4.2 3B

GPT-5.4 (Non-reasoning) vs Granite 4.2 3B. A reliable intelligence-versus-cost verdict requires both models' intelligence scores and benchmark-run costs. Missing data is not a performance verdict.

OpenAIGPT-5.4 (Non-reasoning)Not ranked · benchmark data missing
Intelligence
18.20
Total cost
-
IBMGranite 4.2 3BOn the open-weights frontier
Intelligence
9.10
Total cost
$14.9
Verdict

Comparison incomplete

A reliable intelligence-versus-cost verdict requires both models' intelligence scores and benchmark-run costs. Missing data is not a performance verdict.

MetricGPT-5.4 (Non-reasoning)Granite 4.2 3BDifference
Intelligence (higher is better)18.29.1−50.0% Granite 4.2 3B is 50.0% less intelligent
Total AA cost (lower is better)Not reported$14.91— not comparable
End-to-end time (lower is better)7.59s49.98s+558.5% Granite 4.2 3B is 558.5% slower
Price per 1M tokens (in / out)$2.50 / $15.0$0.03 / $0.12
Input modalitiesText, ImageText
Open this matchup in the interactive pickerGPT-5.4 (Non-reasoning): what beats itGranite 4.2 3B: what beats it
Matchup

Compare something else

Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/gpt-5-4-non-reasoning/granite-4-2-3b · Markdown: /api/v1/compare/gpt-5-4-non-reasoning/granite-4-2-3b?format=md · llms.txt · agent skill