GPT-5.4 Nano (Non-reasoning) — benchmark results

GPT-5.4 Nano evaluated with reasoning disabled. Provider: OpenAI. Released 2026-03-17. Access: API.

Unified ELO 1495 ± 8, rank #816 of 1776 rated models, from 295 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Polish Summarization - PSC25.44Score (%)97.3
EuroEval Lithuanian NLU - Atsiliepimai49.5Sentiment classification Score (%)96
EuroEval Italian NLU - ScaLA IT44.93Linguistic acceptability Score (%)89.6
EuroEval Slovene NLU - Sentinews52.23Sentiment classification Score (%)87.9
EuroEval Portuguese NLU - ScaLA PT35Linguistic acceptability Score (%)87
EuroEval Serbian NLU - MMS SR48.55Sentiment classification Score (%)83.7
EuroEval Dutch Simplification - Duidelijke Taal53.92Score (%)81.8
EuroEval Catalan NLU - Guia CAT68.06Sentiment classification Score (%)81
EuroEval Bulgarian Knowledge55Knowledge Average Score (%)79.9
EuroEval Greek Summarization - Greek Wikipedia31.34Score (%)78.8
EuroEval Faroese NLU - FoSent51.6Sentiment classification Score (%)77.4
EuroEval Spanish NLU - MLQA ES64.06Reading comprehension Score (%)76.4

Interactive version: theaggregate.ai/model?slug=gpt-5-4-nano-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.