{
 "benchmarkVersion": "arcadebench-v0",
 "policyVersion": "joule-capacity-v1",
 "reportVersion": "1.2",
 "generatedFrom": "7 tournament exports dated 2026-08-20",
 "tournaments": 7,
 "series": 73,
 "matches": 265,
 "matchesDrawn": 176,
 "decisionsRecorded": 2167,
 "decisionsScored": 2156,
 "fallbacksExcluded": 11,
 "models": 17,
 "optimalRate": 0.9429,
 "avgRegret": 0.0598,
 "catastrophicErrors": 6,
 "totalJoules": 14289691.5,
 "meanJoulesPerDecision": 6628,
 "joulesOnNonOptimal": 522551.4,
 "overallEfficiency": 0.9634,
 "coinFlips": 19,
 "coinFlipShareOfSeries": 0.2603,
 "grandFinalsByCoinFlip": 1,
 "coinFlipsWhereModelReplied": 0,
 "zeroReasoningDecisions": 601,
 "effortAllocation": {
  "rDifficultyReasoningTokens": -0.0436,
  "rDifficultyAnswerTokens": -0.0261,
  "rPliesReasoningTokens": 0.2301,
  "medianReasoningTokens": 46,
  "n": 2155
 },
 "perModel": [
  {
   "model": "google/gemini-3.1-pro-preview",
   "n": 62,
   "optimalRate": 1.0,
   "avgRegret": 0.0,
   "catastrophicRate": 0.0,
   "totalJoules": 758736.0,
   "joulesPerDecision": 12238,
   "efficiency": 1.0
  },
  {
   "model": "anthropic/claude-fable-5",
   "n": 49,
   "optimalRate": 1.0,
   "avgRegret": 0.0,
   "catastrophicRate": 0.0,
   "totalJoules": 2359620.0,
   "joulesPerDecision": 48156,
   "efficiency": 1.0
  },
  {
   "model": "anthropic/claude-opus-5",
   "n": 39,
   "optimalRate": 1.0,
   "avgRegret": 0.0,
   "catastrophicRate": 0.0,
   "totalJoules": 826866.0,
   "joulesPerDecision": 21202,
   "efficiency": 1.0
  },
  {
   "model": "google/gemini-3.7-flash",
   "n": 238,
   "optimalRate": 1.0,
   "avgRegret": 0.0,
   "catastrophicRate": 0.0,
   "totalJoules": 340958.1,
   "joulesPerDecision": 1433,
   "efficiency": 1.0
  },
  {
   "model": "google/gemma-4-31b-it",
   "n": 83,
   "optimalRate": 1.0,
   "avgRegret": 0.0,
   "catastrophicRate": 0.0,
   "totalJoules": 92444.0,
   "joulesPerDecision": 1114,
   "efficiency": 1.0
  },
  {
   "model": "x-ai/grok-4.6",
   "n": 246,
   "optimalRate": 0.9959,
   "avgRegret": 0.0041,
   "catastrophicRate": 0.0,
   "totalJoules": 2580696.0,
   "joulesPerDecision": 10491,
   "efficiency": 0.9864
  },
  {
   "model": "openai/gpt-5.6-sol",
   "n": 132,
   "optimalRate": 0.9848,
   "avgRegret": 0.0152,
   "catastrophicRate": 0.0,
   "totalJoules": 1446750.0,
   "joulesPerDecision": 10960,
   "efficiency": 0.9738
  },
  {
   "model": "qwen/qwen3.8-27b",
   "n": 70,
   "optimalRate": 0.9714,
   "avgRegret": 0.0286,
   "catastrophicRate": 0.0,
   "totalJoules": 480446.3,
   "joulesPerDecision": 6864,
   "efficiency": 0.9658
  },
  {
   "model": "anthropic/claude-sonnet-5",
   "n": 352,
   "optimalRate": 0.9744,
   "avgRegret": 0.0256,
   "catastrophicRate": 0.0,
   "totalJoules": 3144088.8,
   "joulesPerDecision": 8932,
   "efficiency": 0.964
  },
  {
   "model": "openai/gpt-5.6-terra",
   "n": 133,
   "optimalRate": 0.9398,
   "avgRegret": 0.0602,
   "catastrophicRate": 0.0,
   "totalJoules": 868003.2,
   "joulesPerDecision": 6526,
   "efficiency": 0.932
  },
  {
   "model": "openai/gpt-oss-120b",
   "n": 150,
   "optimalRate": 0.9,
   "avgRegret": 0.1,
   "catastrophicRate": 0.0,
   "totalJoules": 56587.2,
   "joulesPerDecision": 377,
   "efficiency": 0.9174
  },
  {
   "model": "z-ai/glm-5.3",
   "n": 44,
   "optimalRate": 0.9773,
   "avgRegret": 0.0227,
   "catastrophicRate": 0.0,
   "totalJoules": 152039.1,
   "joulesPerDecision": 3455,
   "efficiency": 0.8946
  },
  {
   "model": "meta/muse-spark-1.2",
   "n": 66,
   "optimalRate": 0.9242,
   "avgRegret": 0.0758,
   "catastrophicRate": 0.0,
   "totalJoules": 493511.9,
   "joulesPerDecision": 7477,
   "efficiency": 0.8903
  },
  {
   "model": "openai/gpt-5.6-luna",
   "n": 280,
   "optimalRate": 0.9,
   "avgRegret": 0.1036,
   "catastrophicRate": 0.0036,
   "totalJoules": 263987.2,
   "joulesPerDecision": 943,
   "efficiency": 0.8245
  },
  {
   "model": "google/gemini-3.5-flash-lite",
   "n": 101,
   "optimalRate": 0.8119,
   "avgRegret": 0.1881,
   "catastrophicRate": 0.0,
   "totalJoules": 90848.1,
   "joulesPerDecision": 899,
   "efficiency": 0.7867
  },
  {
   "model": "anthropic/claude-haiku-4.5",
   "n": 83,
   "optimalRate": 0.7349,
   "avgRegret": 0.3012,
   "catastrophicRate": 0.0361,
   "totalJoules": 274496.4,
   "joulesPerDecision": 3307,
   "efficiency": 0.6534
  },
  {
   "model": "thinkingmachines/inkling",
   "n": 28,
   "optimalRate": 0.6071,
   "avgRegret": 0.4643,
   "catastrophicRate": 0.0714,
   "totalJoules": 59613.2,
   "joulesPerDecision": 2129,
   "efficiency": 0.5729
  }
 ]
}