{"model_scores":[{"model_id":"gpt-5.3-codex-xhigh","display_name":"gpt-5.3-codex","lab":"openai","prompt":"bare","release_date":"2026-02-05","precision":"unknown","base_model":"gpt-5.3-codex","reasoning_effort":"xhigh","provider":"openai","notes":null,"pass_rate":0.8138297872340425,"ci_lower":0.7713823080706448,"ci_upper":0.849929312352184,"total_questions":376,"total_runs":376,"cost_avg":0.176053295212766,"cost_total":66.19603900000001,"latency_p50":111224.5,"category_pass_rates":{"calculations":1,"conceptual":1,"consensus-epoch-processing":0.6538461538461539,"consensus-fork-choice":0.9333333333333333,"consensus-shuffling":0,"consensus-state-transition":0.8666666666666667,"constants":1,"cross-fork":1,"eip-interactions":0,"evm-execution":0.7833333333333333,"fork-choice-compound":0.78,"trick":0.6666666666666666}},{"model_id":"gpt-5.3-codex-high","display_name":"gpt-5.3-codex","lab":"openai","prompt":"bare","release_date":"2026-02-05","precision":"unknown","base_model":"gpt-5.3-codex","reasoning_effort":"high","provider":"openai","notes":null,"pass_rate":0.7952127659574468,"ci_lower":0.7515340438320556,"ci_upper":0.8329201075826395,"total_questions":376,"total_runs":376,"cost_avg":0.09932827327127658,"cost_total":37.347430749999994,"latency_p50":44590.5,"category_pass_rates":{"calculations":1,"conceptual":0.8333333333333334,"consensus-epoch-processing":0.6346153846153846,"consensus-fork-choice":0.9333333333333333,"consensus-shuffling":0,"consensus-state-transition":0.8666666666666667,"constants":1,"cross-fork":0.75,"eip-interactions":0,"evm-execution":0.7333333333333333,"fork-choice-compound":0.82,"trick":0.6666666666666666}},{"model_id":"gpt-5.4-xhigh","display_name":"gpt-5.4","lab":"openai","prompt":"bare","release_date":"2026-03-05","precision":"unknown","base_model":"gpt-5.4","reasoning_effort":"xhigh","provider":"openai","notes":null,"pass_rate":0.7686170212765957,"ci_lower":0.7234026820321796,"ci_upper":0.8083979422280383,"total_questions":376,"total_runs":376,"cost_avg":0.17222091090425537,"cost_total":64.75506250000002,"latency_p50":111845.5,"category_pass_rates":{"calculations":1,"conceptual":1,"consensus-epoch-processing":0.6538461538461539,"consensus-fork-choice":0.9333333333333333,"consensus-shuffling":0,"consensus-state-transition":0.8333333333333334,"constants":1,"cross-fork":0.5,"eip-interactions":0,"evm-execution":0.75,"fork-choice-compound":0.6,"trick":0.6666666666666666}},{"model_id":"gpt-5.3-codex-medium","display_name":"gpt-5.3-codex","lab":"openai","prompt":"bare","release_date":"2026-02-05","precision":"unknown","base_model":"gpt-5.3-codex","reasoning_effort":"medium","provider":"openai","notes":null,"pass_rate":0.7074468085106383,"ci_lower":0.6595493030558298,"ci_upper":0.751148208749091,"total_questions":376,"total_runs":376,"cost_avg":0.037499395611702124,"cost_total":14.09977275,"latency_p50":22313.5,"category_pass_rates":{"calculations":1,"conceptual":0.8333333333333334,"consensus-epoch-processing":0.6153846153846154,"consensus-fork-choice":0.8666666666666667,"consensus-shuffling":0,"consensus-state-transition":0.8,"constants":0.9824561403508771,"cross-fork":0.5,"eip-interactions":0,"evm-execution":0.7166666666666667,"fork-choice-compound":0.38,"trick":0.5}}],"categories":[{"name":"calculations","pass_rate":1,"ci_lower":0.6755843804891231,"ci_upper":1,"total":8,"passed":8},{"name":"constants","pass_rate":0.9956,"ci_lower":0.9755781171086069,"ci_upper":0.9992253694097745,"total":228,"passed":227},{"name":"conceptual","pass_rate":0.9167,"ci_lower":0.741507847985494,"ci_upper":0.9768416721570264,"total":24,"passed":22},{"name":"consensus-fork-choice","pass_rate":0.9167,"ci_lower":0.8193083153188752,"ci_upper":0.963880075949514,"total":60,"passed":55},{"name":"consensus-state-transition","pass_rate":0.8417,"ci_lower":0.7901484413732998,"ci_upper":0.8824193239300775,"total":240,"passed":202},{"name":"evm-execution","pass_rate":0.7458,"ci_lower":0.705036603951042,"ci_upper":0.7827263333408321,"total":480,"passed":358},{"name":"cross-fork","pass_rate":0.6875,"ci_lower":0.4440393186673406,"ci_upper":0.8583556494803894,"total":16,"passed":11},{"name":"fork-choice-compound","pass_rate":0.645,"ci_lower":0.5765198448160798,"ci_upper":0.7080148134676073,"total":200,"passed":129},{"name":"consensus-epoch-processing","pass_rate":0.6394,"ci_lower":0.5721841536397051,"ci_upper":0.7016053192494723,"total":208,"passed":133},{"name":"trick","pass_rate":0.625,"ci_lower":0.42709619037907065,"ci_upper":0.7884086656636855,"total":24,"passed":15},{"name":"eip-interactions","pass_rate":0,"ci_lower":0,"ci_upper":0.48990002040399916,"total":4,"passed":0},{"name":"consensus-shuffling","pass_rate":0,"ci_lower":0,"ci_upper":0.2425007574992425,"total":12,"passed":0}],"total_available_questions":376}
