Add executive benchmark README and all-model financial consolidation

This commit is contained in:
Marcos Lohmann
2026-10-15 11:21:53 -03:00
parent 62258836bb
commit 47e241a321
3 changed files with 1021 additions and 1 deletions

View File

@@ -0,0 +1,840 @@
{
"scope": "Todas as 210 chamadas bem-sucedidas com contadores de uso preservados: 3 de validação, 7 de compatibilidade e 200 do benchmark. Erros sem usage ficam fora; não se presume custo zero.",
"valuation_date": "2026-10-15",
"currency": "USD",
"pricing_basis": "Uso medido de 14 e 15/10 reprecificado pela lista Oracle de 15/10/2026, padrão e contexto curto; não é fatura conciliada.",
"sources": {
"paired_benchmark": "results/comparison-2026-10-15/measured-usage.json",
"compatibility_matrix": "../evidence/grok-websearch-models.json",
"poc_validation": "Contadores de três chamadas preservados neste JSON; respostas brutas e inventário original permanecem locais."
},
"pricing": {
"list_url": "https://www.oracle.com/cloud/price-list/",
"data_url": "https://www.oracle.com/a/ocom/docs/pricing/cloud-price-list.json",
"supplement_checked_at": "2026-10-15T14:16:43.202628+00:00",
"supplement_data_sha256": "354aa11836c07e2c065700b0efd80253e8c12d9e643e55160fc1ccd9ea7235d9",
"rates": {
"grok-4.20": {
"input_short": {
"sku": "B111910",
"usd": 1.25
},
"cached_short": {
"sku": "B111911",
"usd": 0.2
},
"output_short": {
"sku": "B112076",
"usd": 2.5
}
},
"grok-4.3": {
"input_short": {
"sku": "B112080",
"usd": 1.25
},
"cached_short": {
"sku": "B112081",
"usd": 0.2
},
"output_short": {
"sku": "B112082",
"usd": 2.5
}
},
"grok-4.6": {
"input_short": {
"sku": "B112086",
"usd": 2
},
"cached_short": {
"sku": "B112087",
"usd": 0.5
},
"output_short": {
"sku": "B112589",
"usd": 6
}
},
"grok-4.7": {
"input_short": {
"sku": "B113303",
"usd": 2
},
"cached_short": {
"sku": "B113304",
"usd": 0.5
},
"output_short": {
"sku": "B113305",
"usd": 6
}
}
},
"web_search": {
"sku": "B112610",
"usd_per_1000": 5
}
},
"total": {
"interactions": 210,
"input_tokens": 6949610,
"cached_input_tokens": 2527488,
"output_tokens": 217197,
"reasoning_tokens": 76530,
"total_tokens": 7166807,
"web_search_calls": 981,
"input_usd": "5.54306200",
"cached_input_usd": "0.5069184",
"output_usd": "0.5502935",
"web_search_usd": "4.905",
"total_usd": "11.50527390",
"mean_usd": "0.05478701857142857142857142857"
},
"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"
}
},
"by_model": {
"xai.grok-4.20-0309-non-reasoning": {
"interactions": 1,
"input_tokens": 10036,
"cached_input_tokens": 1024,
"output_tokens": 196,
"reasoning_tokens": 0,
"total_tokens": 10232,
"web_search_calls": 2,
"input_usd": "0.011265",
"cached_input_usd": "0.0002048",
"output_usd": "0.00049",
"web_search_usd": "0.01",
"total_usd": "0.0219598",
"mean_usd": "0.0219598"
},
"xai.grok-4.20-0309-reasoning": {
"interactions": 1,
"input_tokens": 7712,
"cached_input_tokens": 256,
"output_tokens": 1491,
"reasoning_tokens": 1375,
"total_tokens": 9203,
"web_search_calls": 1,
"input_usd": "0.00932",
"cached_input_usd": "0.0000512",
"output_usd": "0.0037275",
"web_search_usd": "0.005",
"total_usd": "0.0180987",
"mean_usd": "0.0180987"
},
"xai.grok-4.20-non-reasoning": {
"interactions": 101,
"input_tokens": 4826397,
"cached_input_tokens": 2198656,
"output_tokens": 92994,
"reasoning_tokens": 0,
"total_tokens": 4919391,
"web_search_calls": 649,
"input_usd": "3.28467625",
"cached_input_usd": "0.4397312",
"output_usd": "0.2324850",
"web_search_usd": "3.245",
"total_usd": "7.20189245",
"mean_usd": "0.07130586584158415841584158416"
},
"xai.grok-4.20-reasoning": {
"interactions": 1,
"input_tokens": 7882,
"cached_input_tokens": 1024,
"output_tokens": 1531,
"reasoning_tokens": 1329,
"total_tokens": 9413,
"web_search_calls": 2,
"input_usd": "0.0085725",
"cached_input_usd": "0.0002048",
"output_usd": "0.0038275",
"web_search_usd": "0.01",
"total_usd": "0.0226048",
"mean_usd": "0.0226048"
},
"xai.grok-4.3": {
"interactions": 104,
"input_tokens": 2072301,
"cached_input_tokens": 321792,
"output_tokens": 118899,
"reasoning_tokens": 72106,
"total_tokens": 2191200,
"web_search_calls": 323,
"input_usd": "2.18813625",
"cached_input_usd": "0.0643584",
"output_usd": "0.2972475",
"web_search_usd": "1.615",
"total_usd": "4.16474215",
"mean_usd": "0.04004559759615384615384615385"
},
"xai.grok-4.6": {
"interactions": 1,
"input_tokens": 8712,
"cached_input_tokens": 1280,
"output_tokens": 878,
"reasoning_tokens": 732,
"total_tokens": 9590,
"web_search_calls": 1,
"input_usd": "0.014864",
"cached_input_usd": "0.00064",
"output_usd": "0.005268",
"web_search_usd": "0.005",
"total_usd": "0.025772",
"mean_usd": "0.025772"
},
"xai.grok-4.7": {
"interactions": 1,
"input_tokens": 16570,
"cached_input_tokens": 3456,
"output_tokens": 1208,
"reasoning_tokens": 988,
"total_tokens": 17778,
"web_search_calls": 3,
"input_usd": "0.026228",
"cached_input_usd": "0.001728",
"output_usd": "0.007248",
"web_search_usd": "0.015",
"total_usd": "0.050204",
"mean_usd": "0.050204"
}
},
"compatibility_matrix_by_model": {
"xai.grok-4.20-0309-non-reasoning": {
"interactions": 1,
"input_tokens": 10036,
"cached_input_tokens": 1024,
"output_tokens": 196,
"reasoning_tokens": 0,
"total_tokens": 10232,
"web_search_calls": 2,
"input_usd": "0.011265",
"cached_input_usd": "0.0002048",
"output_usd": "0.00049",
"web_search_usd": "0.01",
"total_usd": "0.0219598",
"mean_usd": "0.0219598"
},
"xai.grok-4.20-0309-reasoning": {
"interactions": 1,
"input_tokens": 7712,
"cached_input_tokens": 256,
"output_tokens": 1491,
"reasoning_tokens": 1375,
"total_tokens": 9203,
"web_search_calls": 1,
"input_usd": "0.00932",
"cached_input_usd": "0.0000512",
"output_usd": "0.0037275",
"web_search_usd": "0.005",
"total_usd": "0.0180987",
"mean_usd": "0.0180987"
},
"xai.grok-4.20-non-reasoning": {
"interactions": 1,
"input_tokens": 10944,
"cached_input_tokens": 1024,
"output_tokens": 224,
"reasoning_tokens": 0,
"total_tokens": 11168,
"web_search_calls": 2,
"input_usd": "0.0124",
"cached_input_usd": "0.0002048",
"output_usd": "0.00056",
"web_search_usd": "0.01",
"total_usd": "0.0231648",
"mean_usd": "0.0231648"
},
"xai.grok-4.20-reasoning": {
"interactions": 1,
"input_tokens": 7882,
"cached_input_tokens": 1024,
"output_tokens": 1531,
"reasoning_tokens": 1329,
"total_tokens": 9413,
"web_search_calls": 2,
"input_usd": "0.0085725",
"cached_input_usd": "0.0002048",
"output_usd": "0.0038275",
"web_search_usd": "0.01",
"total_usd": "0.0226048",
"mean_usd": "0.0226048"
},
"xai.grok-4.3": {
"interactions": 1,
"input_tokens": 8092,
"cached_input_tokens": 1280,
"output_tokens": 898,
"reasoning_tokens": 734,
"total_tokens": 8990,
"web_search_calls": 2,
"input_usd": "0.008515",
"cached_input_usd": "0.000256",
"output_usd": "0.002245",
"web_search_usd": "0.01",
"total_usd": "0.021016",
"mean_usd": "0.021016"
},
"xai.grok-4.6": {
"interactions": 1,
"input_tokens": 8712,
"cached_input_tokens": 1280,
"output_tokens": 878,
"reasoning_tokens": 732,
"total_tokens": 9590,
"web_search_calls": 1,
"input_usd": "0.014864",
"cached_input_usd": "0.00064",
"output_usd": "0.005268",
"web_search_usd": "0.005",
"total_usd": "0.025772",
"mean_usd": "0.025772"
},
"xai.grok-4.7": {
"interactions": 1,
"input_tokens": 16570,
"cached_input_tokens": 3456,
"output_tokens": 1208,
"reasoning_tokens": 988,
"total_tokens": 17778,
"web_search_calls": 3,
"input_usd": "0.026228",
"cached_input_usd": "0.001728",
"output_usd": "0.007248",
"web_search_usd": "0.015",
"total_usd": "0.050204",
"mean_usd": "0.050204"
}
},
"historical_usage_records": [
{
"stage": "poc_validation",
"date": "2026-10-14",
"model": "xai.grok-4.3",
"source_file": "route_20231130_actions_v1_responses.json",
"usage": {
"input_tokens": 7607,
"input_tokens_details": {
"cached_tokens": 4864
},
"output_tokens": 1087,
"output_tokens_details": {
"reasoning_tokens": 843
},
"total_tokens": 8694,
"num_sources_used": 0,
"num_server_side_tools_used": 3,
"cost_in_usd_ticks": 221190500,
"server_side_tool_usage_details": {
"web_search_calls": 3,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 3680,
"output_tokens": 369
}
},
"metrics": {
"interactions": 1,
"input_tokens": 7607,
"cached_input_tokens": 4864,
"output_tokens": 1087,
"reasoning_tokens": 843,
"total_tokens": 8694,
"web_search_calls": 3,
"input_usd": "0.00342875",
"cached_input_usd": "0.0009728",
"output_usd": "0.0027175",
"web_search_usd": "0.015",
"total_usd": "0.02211905"
}
},
{
"stage": "poc_validation",
"date": "2026-10-14",
"model": "xai.grok-4.3",
"source_file": "rest-news-today.json",
"usage": {
"input_tokens": 14359,
"input_tokens_details": {
"cached_tokens": 5760
},
"output_tokens": 1185,
"output_tokens_details": {
"reasoning_tokens": 798
},
"total_tokens": 15544,
"num_sources_used": 0,
"num_server_side_tools_used": 3,
"cost_in_usd_ticks": 298632500,
"server_side_tool_usage_details": {
"web_search_calls": 3,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 7565,
"output_tokens": 466
}
},
"metrics": {
"interactions": 1,
"input_tokens": 14359,
"cached_input_tokens": 5760,
"output_tokens": 1185,
"reasoning_tokens": 798,
"total_tokens": 15544,
"web_search_calls": 3,
"input_usd": "0.01074875",
"cached_input_usd": "0.001152",
"output_usd": "0.0029625",
"web_search_usd": "0.015",
"total_usd": "0.02986325"
}
},
{
"stage": "poc_validation",
"date": "2026-10-14",
"model": "xai.grok-4.3",
"source_file": "route_v1_responses.json",
"usage": {
"input_tokens": 9515,
"input_tokens_details": {
"cached_tokens": 2048
},
"output_tokens": 1076,
"output_tokens_details": {
"reasoning_tokens": 729
},
"total_tokens": 10591,
"num_sources_used": 0,
"num_server_side_tools_used": 3,
"cost_in_usd_ticks": 274333500,
"server_side_tool_usage_details": {
"web_search_calls": 3,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 8450,
"output_tokens": 571
}
},
"metrics": {
"interactions": 1,
"input_tokens": 9515,
"cached_input_tokens": 2048,
"output_tokens": 1076,
"reasoning_tokens": 729,
"total_tokens": 10591,
"web_search_calls": 3,
"input_usd": "0.00933375",
"cached_input_usd": "0.0004096",
"output_usd": "0.00269",
"web_search_usd": "0.015",
"total_usd": "0.02743335"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.20-0309-non-reasoning",
"usage": {
"input_tokens": 10036,
"input_tokens_details": {
"cached_tokens": 1024
},
"output_tokens": 196,
"output_tokens_details": {
"reasoning_tokens": 0
},
"total_tokens": 10232,
"num_sources_used": 0,
"num_server_side_tools_used": 2,
"cost_in_usd_ticks": 219598000,
"server_side_tool_usage_details": {
"web_search_calls": 2,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 9058,
"output_tokens": 138
}
},
"metrics": {
"interactions": 1,
"input_tokens": 10036,
"cached_input_tokens": 1024,
"output_tokens": 196,
"reasoning_tokens": 0,
"total_tokens": 10232,
"web_search_calls": 2,
"input_usd": "0.011265",
"cached_input_usd": "0.0002048",
"output_usd": "0.00049",
"web_search_usd": "0.01",
"total_usd": "0.0219598"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.20-0309-reasoning",
"usage": {
"input_tokens": 7712,
"input_tokens_details": {
"cached_tokens": 256
},
"output_tokens": 1491,
"output_tokens_details": {
"reasoning_tokens": 1375
},
"total_tokens": 9203,
"num_sources_used": 0,
"num_server_side_tools_used": 1,
"cost_in_usd_ticks": 180987000,
"server_side_tool_usage_details": {
"web_search_calls": 1,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 6732,
"output_tokens": 864
}
},
"metrics": {
"interactions": 1,
"input_tokens": 7712,
"cached_input_tokens": 256,
"output_tokens": 1491,
"reasoning_tokens": 1375,
"total_tokens": 9203,
"web_search_calls": 1,
"input_usd": "0.00932",
"cached_input_usd": "0.0000512",
"output_usd": "0.0037275",
"web_search_usd": "0.005",
"total_usd": "0.0180987"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.20-non-reasoning",
"usage": {
"input_tokens": 10944,
"input_tokens_details": {
"cached_tokens": 1024
},
"output_tokens": 224,
"output_tokens_details": {
"reasoning_tokens": 0
},
"total_tokens": 11168,
"num_sources_used": 0,
"num_server_side_tools_used": 2,
"cost_in_usd_ticks": 231648000,
"server_side_tool_usage_details": {
"web_search_calls": 2,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 9966,
"output_tokens": 153
}
},
"metrics": {
"interactions": 1,
"input_tokens": 10944,
"cached_input_tokens": 1024,
"output_tokens": 224,
"reasoning_tokens": 0,
"total_tokens": 11168,
"web_search_calls": 2,
"input_usd": "0.0124",
"cached_input_usd": "0.0002048",
"output_usd": "0.00056",
"web_search_usd": "0.01",
"total_usd": "0.0231648"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.20-reasoning",
"usage": {
"input_tokens": 7882,
"input_tokens_details": {
"cached_tokens": 1024
},
"output_tokens": 1531,
"output_tokens_details": {
"reasoning_tokens": 1329
},
"total_tokens": 9413,
"num_sources_used": 0,
"num_server_side_tools_used": 2,
"cost_in_usd_ticks": 226048000,
"server_side_tool_usage_details": {
"web_search_calls": 2,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 6902,
"output_tokens": 732
}
},
"metrics": {
"interactions": 1,
"input_tokens": 7882,
"cached_input_tokens": 1024,
"output_tokens": 1531,
"reasoning_tokens": 1329,
"total_tokens": 9413,
"web_search_calls": 2,
"input_usd": "0.0085725",
"cached_input_usd": "0.0002048",
"output_usd": "0.0038275",
"web_search_usd": "0.01",
"total_usd": "0.0226048"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.3",
"usage": {
"input_tokens": 8092,
"input_tokens_details": {
"cached_tokens": 1280
},
"output_tokens": 898,
"output_tokens_details": {
"reasoning_tokens": 734
},
"total_tokens": 8990,
"num_sources_used": 0,
"num_server_side_tools_used": 2,
"cost_in_usd_ticks": 210160000,
"server_side_tool_usage_details": {
"web_search_calls": 2,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 6926,
"output_tokens": 456
}
},
"metrics": {
"interactions": 1,
"input_tokens": 8092,
"cached_input_tokens": 1280,
"output_tokens": 898,
"reasoning_tokens": 734,
"total_tokens": 8990,
"web_search_calls": 2,
"input_usd": "0.008515",
"cached_input_usd": "0.000256",
"output_usd": "0.002245",
"web_search_usd": "0.01",
"total_usd": "0.021016"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.6",
"usage": {
"input_tokens": 8712,
"input_tokens_details": {
"cached_tokens": 1280
},
"output_tokens": 878,
"output_tokens_details": {
"reasoning_tokens": 732
},
"total_tokens": 9590,
"num_sources_used": 0,
"num_server_side_tools_used": 1,
"cost_in_usd_ticks": 257720000,
"server_side_tool_usage_details": {
"web_search_calls": 1,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 7510,
"output_tokens": 535
}
},
"metrics": {
"interactions": 1,
"input_tokens": 8712,
"cached_input_tokens": 1280,
"output_tokens": 878,
"reasoning_tokens": 732,
"total_tokens": 9590,
"web_search_calls": 1,
"input_usd": "0.014864",
"cached_input_usd": "0.00064",
"output_usd": "0.005268",
"web_search_usd": "0.005",
"total_usd": "0.025772"
}
},
{
"stage": "compatibility_matrix",
"date": "2026-10-14",
"model": "xai.grok-4.7",
"usage": {
"input_tokens": 16570,
"input_tokens_details": {
"cached_tokens": 3456
},
"output_tokens": 1208,
"output_tokens_details": {
"reasoning_tokens": 988
},
"total_tokens": 17778,
"num_sources_used": 0,
"num_server_side_tools_used": 3,
"cost_in_usd_ticks": 502040000,
"server_side_tool_usage_details": {
"web_search_calls": 3,
"x_search_calls": 0,
"x_posts_fetched": 0,
"x_users_fetched": 0,
"code_interpreter_calls": 0,
"file_search_calls": 0,
"mcp_calls": 0,
"document_search_calls": 0,
"image_generation_calls": 0
},
"context_details": {
"input_tokens": 14183,
"output_tokens": 1022
}
},
"metrics": {
"interactions": 1,
"input_tokens": 16570,
"cached_input_tokens": 3456,
"output_tokens": 1208,
"reasoning_tokens": 988,
"total_tokens": 17778,
"web_search_calls": 3,
"input_usd": "0.026228",
"cached_input_usd": "0.001728",
"output_usd": "0.007248",
"web_search_usd": "0.015",
"total_usd": "0.050204"
}
}
]
}