Home Model series GPT Sol GPT-6.2 Sol GPT-6.2 Sol · Terminal-Bench-Science · 0.1 · Scores and configurations GPT-6.2 Sol · Terminal-Bench-Science · 0.1 · Scores and configurations GPT-6.2 Sol record for Terminal-Bench-Science · 0.1. 5 score records for this release, including 5 exact values (may include historical records). Includes 5 source locations.
Scores and evaluation conditions GPT-6.2 Sol · Terminal-Bench-Science scores and configurations Metric and score Configuration and evaluation conditions Sources Score 43.4% Reported 43.4% · Unit % Published Unconfirmed
GPT-6.2 Sol low Settings GPT-6.2 Sol; reasoning effort low, explicitly labeled in the official chart data. Unknowns 文章未注明逐项采样设置、完整执行框架和工具配置;按图表所示条件记录。 Tools Tool access not recorded Reasoning budget low Framework not recorded Protocol not specified Conditions GPT-6.2 Sol; reasoning effort low. 官方图表单任务成本:$0.86。逐项采样及完整 harness 未注明。 View this score in rankings Score 63.1% Reported 63.1% · Unit % Published Unconfirmed
GPT-6.2 Sol medium Settings GPT-6.2 Sol; reasoning effort medium, explicitly labeled in the official chart data. Unknowns 文章未注明逐项采样设置、完整执行框架和工具配置;按图表所示条件记录。 Tools Tool access not recorded Reasoning budget medium Framework not recorded Protocol not specified Conditions GPT-6.2 Sol; reasoning effort medium. 官方图表单任务成本:$1.34。逐项采样及完整 harness 未注明。 View this score in rankings Score 67.7% Reported 67.7% · Unit % Published Unconfirmed
GPT-6.2 Sol high Settings GPT-6.2 Sol; reasoning effort high, explicitly labeled in the official chart data. Unknowns 文章未注明逐项采样设置、完整执行框架和工具配置;按图表所示条件记录。 Tools Tool access not recorded Reasoning budget high Framework not recorded Protocol not specified Conditions GPT-6.2 Sol; reasoning effort high. 官方图表单任务成本:$1.51。逐项采样及完整 harness 未注明。 View this score in rankings Score 64.3% Reported 64.3% · Unit % Published Unconfirmed
GPT-6.2 Sol xhigh Settings GPT-6.2 Sol; reasoning effort xhigh, explicitly labeled in the official chart data. Unknowns 文章未注明逐项采样设置、完整执行框架和工具配置;按图表所示条件记录。 Tools Tool access not recorded Reasoning budget xhigh Framework not recorded Protocol not specified Conditions GPT-6.2 Sol; reasoning effort xhigh. 官方图表单任务成本:$1.61。逐项采样及完整 harness 未注明。 View this score in rankings Score 68% Reported 68.0% · Unit % Published Unconfirmed
GPT-6.2 Sol max Settings GPT-6.2 Sol; reasoning effort max, explicitly labeled in the official chart data. Unknowns 文章未注明逐项采样设置、完整执行框架和工具配置;按图表所示条件记录。 Tools Tool access not recorded Reasoning budget max Framework not recorded Protocol not specified Conditions GPT-6.2 Sol; reasoning effort max. 官方图表单任务成本:$2.21。逐项采样及完整 harness 未注明。 View this score in rankings
Benchmark versions and metrics Aliases Terminal-Bench-Science 0.1
Score · % · Higher is better