Home Model series GPT unsuffixed models GPT-5.5 GeneBench GPT-5.5 · GeneBench (2026-04-23 report; 103 evaluations across 10 domains) 33.2% · Main text evidence (ev-gpt55-0032) GPT-5.5 · GeneBench (2026-04-23 report; 103 evaluations across 10 domains) 33.2% · Main text evidence (ev-gpt55-0032) Official source for GeneBench · 2026-04-23 report; 103 evaluations across 10 domains in GPT-5.5. GPT-5.5 Pro · 33.2%. See “Introducing GPT-5.5 | OpenAI” at Main article / Table 5 / row “GeneBench” / column “GPT-5.5 Pro”. Evidence ev-gpt55-0032.
Source excerpt and location GPT-5.5 · GeneBench (2026-04-23 report; 103 evaluations across 10 domains), configuration GPT-5.5 Pro, Mean pass rate 33.2% (exact score publication date unconfirmed). See “Introducing GPT-5.5 | OpenAI” at Main article / Table 5 / row “GeneBench” / column “GPT-5.5 Pro”. View score conditions
GeneBench | GPT-5.5 Pro: 33.2% Official material Introducing GPT-5.5 | OpenAI Location Main text · Main article / Table 5 / row “GeneBench” / column “GPT-5.5 Pro” Material version 官方英文主文全文及可见表格;原始发布日 2026-04-23;当前页面修订时间未知 Version published Unconfirmed Disclosure date Unconfirmed · Launch or later disclosure unconfirmed Collected 2026-11-09T21:24:52.903+00:00 Historical basis No reliable historical version evidence available Citation text GPT-5.5 · GeneBench (2026-04-23 report; 103 evaluations across 10 domains), configuration GPT-5.5 Pro, Mean pass rate 33.2% (exact score publication date unconfirmed). Official source: Introducing GPT-5.5 | OpenAI (Main text, 官方英文主文全文及可见表格;原始发布日 2026-04-23;当前页面修订时间未知), Main article / Table 5 / row “GeneBench” / column “GPT-5.5 Pro”, https://openai.com/index/introducing-gpt-5-5/. Benchlog https://benchlog.lekshome.cc/evidence/gpt-5-5/ev-gpt55-0032/?lang=en
Related scores and conditions GPT-5.5 · GeneBench scores and configurations Metric and score Configuration and evaluation conditions Evidence relation Mean pass rate 33.2% Reported 33.2% · Unit % Published Unconfirmed
GPT-5.5 Pro Settings Same underlying model with a setting that uses parallel test-time compute. Unknowns Parallel compute budget and exact Pro harness settings are not stated. Tools Tools allowed Reasoning budget not recorded Framework Pro harness Protocol GeneBench full suite; separately reported GPT Pro harness; mean unweighted per-problem pass rate over repeated runs; 95% CI in source excerpt Conditions GPT-5.5 Pro; parallel test-time compute; exact compute budget unspecified.; GPT-5.5 Pro uses parallel test-time compute; report separates Pro-harness runs; compute budget unspecified. Directly supports this score
View this score in rankings