GPT-5.4 Mini (High) — benchmark results

GPT-5.4 Mini evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2026-03-17. Access: API.

Unified ELO 1671 ± 11, rank #318 of 1776 rated models, from 308 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Bulgarian NLU73.19NLU Average Score (%)100
EuroEval Latvian NLU - Latvian Twitter Sentiment56.55Sentiment classification Score (%)100
EuroEval Polish NLU - KPWr NER78.64Named entity recognition Score (%)100
EuroEval Swedish Knowledge88.01Knowledge Average Score (%)100
EuroEval Swedish NLU - SUC383.44Named entity recognition Score (%)100
EuroEval Ukrainian NLU - NER UK83.05Named entity recognition Score (%)100
EuroEval German NLU - ScaLA DE72.72Linguistic acceptability Score (%)99.7
EuroEval Czech NLU62.29NLU Average Score (%)99.6
EuroEval Bulgarian NLU - BG NER Bsnlp92.54Named entity recognition Score (%)99.5
EuroEval Catalan NLU - Guia CAT76.13Sentiment classification Score (%)99.5
EuroEval Slovene NLU - SSJ500K NER80.44Named entity recognition Score (%)99.5
EuroEval German NLU63.89NLU Average Score (%)99.4

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