Home Model series Claude Sonnet Claude Sonnet 5.5 Claude Sonnet 5.5 · PhysicianBench · 100 physician tasks in an electronic health record; pass rate across attempts · Scores and configurations Claude Sonnet 5.5 · PhysicianBench · 100 physician tasks in an electronic health record; pass rate across attempts · Scores and configurations Claude Sonnet 5.5 record for PhysicianBench · 100 physician tasks in an electronic health record; pass rate across attempts. 5 score records for this release, including 5 exact values (may include historical records). Includes 2 source locations.
Scores and evaluation conditions Claude Sonnet 5.5 · PhysicianBench scores and configurations Metric and score Configuration and evaluation conditions Sources Pass rate 63.2% Reported 63.2% · Unit % Published Unconfirmed
Max effort Settings Max reasoning effort. Unknowns Adaptive thinking status is benchmark-specific. Tools Tool access not recorded Reasoning budget max Framework not recorded Protocol PhysicianBench, 100 physician EHR tasks. Conditions Max effort; same harness across models; Opus 5 grader. View this score in rankings Pass rate 27.2% Reported 27.2% at Low · Unit % Published Unconfirmed
Low effort Settings Low reasoning effort. Unknowns Sampling settings not stated unless recorded per result. Tools Tool access not recorded Reasoning budget low Framework not recorded Protocol PhysicianBench, 100 physician EHR tasks. Conditions Low effort. View this score in rankings Pass rate 30% Reported 30.0% at Medium · Unit % Published Unconfirmed
Medium effort Settings Medium reasoning effort. Unknowns Sampling settings not stated unless recorded per result. Tools Tool access not recorded Reasoning budget medium Framework not recorded Protocol PhysicianBench, 100 physician EHR tasks. Conditions Medium effort. View this score in rankings Pass rate 47.6% Reported 47.6% · Unit % Published Unconfirmed
High effort Settings High reasoning effort. Unknowns Sampling settings not stated unless recorded per result. Tools Tool access not recorded Reasoning budget high Framework not recorded Protocol PhysicianBench, 100 physician EHR tasks. Conditions Figure labels the effort High. All models used safety classifiers and the same harness; Claude Opus 5 graded responses. View this score in rankings Pass rate 56.4% Reported 56.4% · Unit % Published Unconfirmed
Xhigh effort Settings Xhigh reasoning effort. Unknowns Sampling settings not stated unless recorded per result. Tools Tool access not recorded Reasoning budget xhigh Framework not recorded Protocol PhysicianBench, 100 physician EHR tasks. Conditions Figure labels the effort Xhigh. All models used safety classifiers and the same harness; Claude Opus 5 graded responses. View this score in rankings
Benchmark versions and metrics Aliases PhysicianBench
Pass rate · % · Higher is better