GPT-5 Mini (Medium) — benchmark results

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

Unified ELO 1752 ± 17, rank #194 of 1776 rated models, from 84 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Elimination Game (Lechmazur)5.73TrueSkill μ96.6
SLR-Bench - Easy98Accuracy (%)95.5
MATH Level 596.79Accuracy (%)92.6
SLR-Bench - Hard23Accuracy (%)91.8
SLR-Bench - Medium50Accuracy (%)89.1
IneqMath30.5Overall Accuracy (self-reported)87.8
AA IFBench71.16Accuracy (%)86.6
AI for Education Pedagogy - Technology84.91Accuracy (%)86.4
SLR-Bench - Basic100Accuracy (%)86.4
AI for Education Visual Maths - Algebra92.31Accuracy (%)85.8
AI for Education Visual Maths - Measurement91.89Accuracy (%)85.8
FormationEval96.4Accuracy (%)85.2

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