Home Model series GPT mini GPT mini · Humanity's Last Exam · 2,500-question benchmark · Release history GPT mini · Humanity's Last Exam · 2,500-question benchmark · Release history View scores, configurations, and disclosures for Humanity's Last Exam · 2,500-question benchmark across 2 GPT mini releases, ordered by release date.
Differences are calculated only for unique, active, exact values between direct predecessors when the unit, metric, and explicitly recorded configuration and conditions match.
2025-08-07 · 1 score record for this release, including 1 exact value (may include historical records)
GPT-5 mini · Humanity's Last Exam scores and configurations Metric and score Configuration and evaluation conditions Sources Accuracy 16.7% Reported 16.7% · Unit % Published 2025-08-07
GPT-5 mini (high) Settings API mini model; benchmark table labels high reasoning effort. Unknowns none recorded Tools Tools forbidden Reasoning budget high Framework not recorded Protocol HLE (no tools) Conditions API GPT-5 mini; high reasoning effort. View this score in rankings
Detailed benchmarks / HLE (no tools) Main text · Introducing GPT-5 for developers | OpenAI 2026-03-17 · 2 score records for this release, including 2 exact values (may include historical records)
Records for the direct predecessor do not provide matching conditions for a valid difference.
GPT-5.4 mini · Humanity's Last Exam scores and configurations Metric and score Configuration and evaluation conditions Sources Accuracy 41.5% Reported 41.5% · Unit % Published Unconfirmed
GPT-5.4 mini (xhigh) Settings The official comparison table labels GPT-5.4 mini (xhigh). Unknowns none recorded Tools Tools allowed Reasoning budget xhigh Framework not recorded Protocol Humanity's Last Exam; with tools. Conditions Official comparison table labels GPT-5.4 mini (xhigh). View this score in rankings Accuracy 28.2% Reported 28.2% · Unit % Published Unconfirmed
GPT-5.4 mini (xhigh) Settings The official comparison table labels GPT-5.4 mini (xhigh). Unknowns none recorded Tools Tools forbidden Reasoning budget xhigh Framework not recorded Protocol Humanity's Last Exam; without tools. Conditions Official comparison table labels GPT-5.4 mini (xhigh). View this score in rankings
Intelligence / HLE w/ tool Main text · GPT-5.4 mini 与 nano 正式发布 | OpenAI Intelligence / HLE w/o tools Main text · GPT-5.4 mini 与 nano 正式发布 | OpenAI