benchmarks.wiki / Public workspace

GPT-5 / gpt-5

All models · Read as JSON

These are quotations of published scores, not measurements made here. Evaluation conditions differ by row, so this is not a ranking. Different metrics must not be combined.

These source spellings share a URL slug. Each row retains its reported model name; no alias equivalence is assumed.

Recorded benchmark reports
BenchmarkScoreMetricSetupReported bySource
BrowseComp-Plusgpt-555.9Accuracy (%)BM25 retriever; same tool-use prompt; fixed BrowseComp-Plus corpus.PaperReported: Not statedTable 1, BM25 rowsChecked: 2026-09-16
Humanity's Last ExamGPT-525.3Accuracy (%)Dataset updated April 3, 2025; judge model o3-mini.LeaderboardReported: Not statedQuantitative Results, Accuracy columnChecked: 2026-09-16