GPT-5 Mini (High) — benchmark results

GPT-5 Mini evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.

Unified ELO 1714 ± 15, rank #251 of 1776 rated models, from 236 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM2014 Code 2025-0975.09Median Score100
LLM2014 Code 2025-09 - C++7.29Score100
LLM2014 Code 2025-09 - Golang8.4Score100
LLM2014 Code 2025-09 - Java9Score100
LLM2014 Code 2025-11 - Golang9.2Score100
EuroEval German NLU - Sb10K64.63Sentiment classification Score (%)99.7
SLR-Bench - Easy100Accuracy (%)99.1
EuroEval Portuguese Knowledge90.72Knowledge Average Score (%)98.9
EuroEval French Knowledge85.28Knowledge Average Score (%)98.5
EuroEval Italian Knowledge87.24Knowledge Average Score (%)98.2
SLR-Bench - Medium74Accuracy (%)98.2
LLM2014 Logic 2025-1079.37Median Score98

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