Files
poc-grok-websearch/benchmarks/screening/financial-consolidation.json

204 lines
6.5 KiB
JSON

{
"scope": "294 interações com custo conhecido: histórico original de 210 + auditoria retrospectiva de 84. As etapas são diferentes e não formam uma amostra única para ranking.",
"valuation_date": "2026-10-15",
"currency": "USD",
"pricing_basis": "Uso medido a preços de lista Oracle; não é fatura conciliada. Faixas de contexto longo resolvidas conforme adendo.",
"sources": {
"baseline": "../financial-consolidation.json",
"screening": "results/screening-2026-10-15-v1/measured-usage.json",
"pricing": "oracle-prices.json",
"accounting_addendum": "ADENDO-CONTABILIDADE.md"
},
"source_sha256": {
"baseline": "f21cc006fbdab19431617c3a277a2a9a7a97a98ba8e8e74dcb3ca5aaff05958f",
"screening": "5282c1859b63a1ae18b6162ec9b1c378f4971af8a431edfce16cc223907b18f8"
},
"screening_http_attempts": 91,
"screening_infrastructure_rejections_without_usage": 7,
"unknown_cost_note": "Rejeições sem usage não foram valoradas como zero; estão fora do total conhecido.",
"total": {
"interactions": 294,
"input_tokens": 12046625,
"cached_input_tokens": 4997248,
"output_tokens": 412483,
"reasoning_tokens": 208531,
"total_tokens": 12459108,
"web_search_calls": 1519,
"input_usd": "10.11902675",
"cached_input_usd": "1.7459200",
"output_usd": "1.5034960",
"web_search_usd": "7.595",
"total_usd": "20.96344275",
"mean_usd": "0.07130422704081632653061224490"
},
"by_stage": {
"poc_validation": {
"interactions": 3,
"input_tokens": 31481,
"cached_input_tokens": 12672,
"output_tokens": 3348,
"reasoning_tokens": 2370,
"total_tokens": 34829,
"web_search_calls": 9,
"input_usd": "0.02351125",
"cached_input_usd": "0.0025344",
"output_usd": "0.0083700",
"web_search_usd": "0.045",
"total_usd": "0.07941565",
"mean_usd": "0.02647188333333333333333333333"
},
"compatibility_matrix": {
"interactions": 7,
"input_tokens": 69948,
"cached_input_tokens": 9344,
"output_tokens": 6426,
"reasoning_tokens": 5158,
"total_tokens": 76374,
"web_search_calls": 13,
"input_usd": "0.0911645",
"cached_input_usd": "0.0032896",
"output_usd": "0.0233660",
"web_search_usd": "0.065",
"total_usd": "0.1828201",
"mean_usd": "0.02611715714285714285714285714"
},
"paired_benchmark": {
"interactions": 200,
"input_tokens": 6848181,
"cached_input_tokens": 2505472,
"output_tokens": 207423,
"reasoning_tokens": 69002,
"total_tokens": 7055604,
"web_search_calls": 959,
"input_usd": "5.42838625",
"cached_input_usd": "0.5010944",
"output_usd": "0.5185575",
"web_search_usd": "4.795",
"total_usd": "11.24303815",
"mean_usd": "0.05621519075"
},
"retrospective_screening": {
"interactions": 84,
"input_tokens": 5097015,
"cached_input_tokens": 2469760,
"output_tokens": 195286,
"reasoning_tokens": 132001,
"total_tokens": 5292301,
"web_search_calls": 538,
"input_usd": "4.57596475",
"cached_input_usd": "1.2390016",
"output_usd": "0.9532025",
"web_search_usd": "2.690",
"total_usd": "9.45816885",
"mean_usd": "0.1125972482142857142857142857"
}
},
"by_model": {
"xai.grok-4.20-0309-non-reasoning": {
"interactions": 13,
"input_tokens": 586469,
"cached_input_tokens": 269312,
"output_tokens": 12464,
"reasoning_tokens": 0,
"total_tokens": 598933,
"web_search_calls": 83,
"input_usd": "0.39644625",
"cached_input_usd": "0.0538624",
"output_usd": "0.0311600",
"web_search_usd": "0.415",
"total_usd": "0.89646865",
"mean_usd": "0.06895912692307692307692307692"
},
"xai.grok-4.20-0309-reasoning": {
"interactions": 13,
"input_tokens": 493766,
"cached_input_tokens": 188928,
"output_tokens": 34980,
"reasoning_tokens": 25922,
"total_tokens": 528746,
"web_search_calls": 61,
"input_usd": "0.38104750",
"cached_input_usd": "0.0377856",
"output_usd": "0.0874500",
"web_search_usd": "0.305",
"total_usd": "0.81128310",
"mean_usd": "0.06240639230769230769230769231"
},
"xai.grok-4.20-non-reasoning": {
"interactions": 113,
"input_tokens": 5425184,
"cached_input_tokens": 2451328,
"output_tokens": 104333,
"reasoning_tokens": 0,
"total_tokens": 5529517,
"web_search_calls": 728,
"input_usd": "3.71732000",
"cached_input_usd": "0.4902656",
"output_usd": "0.2608325",
"web_search_usd": "3.640",
"total_usd": "8.10841810",
"mean_usd": "0.07175591238938053097345132743"
},
"xai.grok-4.20-reasoning": {
"interactions": 13,
"input_tokens": 423457,
"cached_input_tokens": 139392,
"output_tokens": 31739,
"reasoning_tokens": 23469,
"total_tokens": 455196,
"web_search_calls": 58,
"input_usd": "0.35508125",
"cached_input_usd": "0.0278784",
"output_usd": "0.0793475",
"web_search_usd": "0.290",
"total_usd": "0.75230715",
"mean_usd": "0.05786978076923076923076923077"
},
"xai.grok-4.3": {
"interactions": 116,
"input_tokens": 2323819,
"cached_input_tokens": 354560,
"output_tokens": 132436,
"reasoning_tokens": 80151,
"total_tokens": 2456255,
"web_search_calls": 362,
"input_usd": "2.46157375",
"cached_input_usd": "0.0709120",
"output_usd": "0.3310900",
"web_search_usd": "1.810",
"total_usd": "4.67357575",
"mean_usd": "0.04028944612068965517241379310"
},
"xai.grok-4.6": {
"interactions": 13,
"input_tokens": 981041,
"cached_input_tokens": 449024,
"output_tokens": 43216,
"reasoning_tokens": 35777,
"total_tokens": 1024257,
"web_search_calls": 97,
"input_usd": "1.064034",
"cached_input_usd": "0.224512",
"output_usd": "0.259296",
"web_search_usd": "0.485",
"total_usd": "2.032842",
"mean_usd": "0.1563724615384615384615384615"
},
"xai.grok-4.7": {
"interactions": 13,
"input_tokens": 1812889,
"cached_input_tokens": 1144704,
"output_tokens": 53315,
"reasoning_tokens": 43212,
"total_tokens": 1866204,
"web_search_calls": 130,
"input_usd": "1.743524",
"cached_input_usd": "0.840704",
"output_usd": "0.454320",
"web_search_usd": "0.650",
"total_usd": "3.688548",
"mean_usd": "0.2837344615384615384615384615"
}
}
}