{"id":"W4412791697","doi":"10.2196/76925","title":"Performance Evaluation of 18 Generative AI Models (ChatGPT, Gemini, Claude, and Perplexity) in 2024 Japanese Pharmacist Licensing Examination: Comparative Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Perplexity; Preprint; Generative grammar; Psychology; Medicine; Computer science; Artificial intelligence; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008432663,0.002471948,0.001555502,0.00318167,0.0007716634,0.002876658,0.001731331,0.00192302,0.003318022],"category_scores_gemma":[0.04079252,0.000549657,0.001664142,0.001453035,0.0006545357,0.002633495,0.002696448,0.00179107,0.002297202],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002881029,"about_ca_system_score_gemma":0.001916622,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01920954,"about_ca_topic_score_gemma":0.01114598,"domain_scores_codex":[0.9927638,0.004037205,0.0005595413,0.001398156,0.0009273359,0.0003139649],"domain_scores_gemma":[0.9627272,0.02948249,0.00104755,0.001545863,0.003498924,0.001698017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.005360623,0.002418421,0.3052346,0.002083655,0.001212997,0.0006572716,0.005143971,0.1061169,0.007230971,0.001525826,0.02457703,0.5384377],"study_design_scores_gemma":[0.000201401,0.001289199,0.08302564,0.0001947649,0.0006388442,0.000376014,0.002342531,0.8953888,0.009225049,0.001946301,0.005169873,0.0002015905],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9643904,0.001760085,0.01833102,0.0008330409,0.0002458109,0.0003316197,0.002827261,0.004289401,0.006991389],"genre_scores_gemma":[0.9773604,0.0004230977,0.01343446,0.0001676064,0.00005959005,0.0002693721,0.006048077,0.0001577392,0.002079697],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01920954,"threshold_uncertainty_score":0.04459667,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2598460124644382,"score_gpt":0.5373930788811854,"score_spread":0.2775470664167472,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}