{
  "apiVersion": "1",
  "methodologyVersion": "2026-09-30-arc-predictive-puzzles",
  "updatedAt": "2026-10-13T12:11:32.230Z",
  "rankings": [
    {
      "id": "composite-iq",
      "rankingName": "Composite IQ",
      "rankingType": "derived",
      "dimension": null,
      "direction": "higher_is_better",
      "modelCount": 164,
      "url": "https://www.aiiq.org/api/rankings/composite-iq"
    },
    {
      "id": "effective-cost",
      "rankingName": "Effective Cost",
      "rankingType": "derived",
      "dimension": null,
      "direction": "lower_is_better",
      "modelCount": 149,
      "url": "https://www.aiiq.org/api/rankings/effective-cost"
    },
    {
      "id": "abstract-reasoning-iq",
      "rankingName": "Abstract Reasoning IQ",
      "rankingType": "dimension",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 181,
      "url": "https://www.aiiq.org/api/rankings/abstract-reasoning-iq"
    },
    {
      "id": "mathematical-reasoning-iq",
      "rankingName": "Mathematical Reasoning IQ",
      "rankingType": "dimension",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 164,
      "url": "https://www.aiiq.org/api/rankings/mathematical-reasoning-iq"
    },
    {
      "id": "academic-reasoning-iq",
      "rankingName": "Academic Reasoning IQ",
      "rankingType": "dimension",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 181,
      "url": "https://www.aiiq.org/api/rankings/academic-reasoning-iq"
    },
    {
      "id": "programmatic-reasoning-iq",
      "rankingName": "Programmatic Reasoning IQ",
      "rankingType": "dimension",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 165,
      "url": "https://www.aiiq.org/api/rankings/programmatic-reasoning-iq"
    },
    {
      "id": "computer-use-iq",
      "rankingName": "Computer Use IQ",
      "rankingType": "dimension",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 164,
      "url": "https://www.aiiq.org/api/rankings/computer-use-iq"
    },
    {
      "id": "reliability-iq",
      "rankingName": "Reliability IQ",
      "rankingType": "dimension",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 164,
      "url": "https://www.aiiq.org/api/rankings/reliability-iq"
    },
    {
      "id": "knowledge-work-iq",
      "rankingName": "Knowledge Work IQ",
      "rankingType": "dimension",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 164,
      "url": "https://www.aiiq.org/api/rankings/knowledge-work-iq"
    },
    {
      "id": "aa-analyst-agent",
      "rankingName": "AA-AnalystAgent",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 23,
      "url": "https://www.aiiq.org/api/rankings/aa-analyst-agent"
    },
    {
      "id": "aa-briefcase",
      "rankingName": "AA-Briefcase v1.1 Elo",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 100,
      "url": "https://www.aiiq.org/api/rankings/aa-briefcase"
    },
    {
      "id": "aa-lcr-v1-1",
      "rankingName": "AA Long Context Reasoning v1.1",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 144,
      "url": "https://www.aiiq.org/api/rankings/aa-lcr-v1-1"
    },
    {
      "id": "aa-omniscience",
      "rankingName": "AA Omniscience",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 148,
      "url": "https://www.aiiq.org/api/rankings/aa-omniscience"
    },
    {
      "id": "agent-arena",
      "rankingName": "Arena.ai Agent Arena",
      "rankingType": "benchmark",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 59,
      "url": "https://www.aiiq.org/api/rankings/agent-arena"
    },
    {
      "id": "agents-last-exam",
      "rankingName": "Agents' Last Exam",
      "rankingType": "benchmark",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 29,
      "url": "https://www.aiiq.org/api/rankings/agents-last-exam"
    },
    {
      "id": "aime",
      "rankingName": "AIME",
      "rankingType": "benchmark",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 56,
      "url": "https://www.aiiq.org/api/rankings/aime"
    },
    {
      "id": "arc-agi-1",
      "rankingName": "ARC-AGI-1",
      "rankingType": "benchmark",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 61,
      "url": "https://www.aiiq.org/api/rankings/arc-agi-1"
    },
    {
      "id": "arc-agi-2",
      "rankingName": "ARC-AGI-2",
      "rankingType": "benchmark",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 61,
      "url": "https://www.aiiq.org/api/rankings/arc-agi-2"
    },
    {
      "id": "arc-agi-3",
      "rankingName": "ARC-AGI-3",
      "rankingType": "benchmark",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 19,
      "url": "https://www.aiiq.org/api/rankings/arc-agi-3"
    },
    {
      "id": "automationbench-aa",
      "rankingName": "AutomationBench-AA",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 97,
      "url": "https://www.aiiq.org/api/rankings/automationbench-aa"
    },
    {
      "id": "browsecomp",
      "rankingName": "BrowseComp",
      "rankingType": "benchmark",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 39,
      "url": "https://www.aiiq.org/api/rankings/browsecomp"
    },
    {
      "id": "bullshitbench-v2",
      "rankingName": "BullshitBench v2",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 81,
      "url": "https://www.aiiq.org/api/rankings/bullshitbench-v2"
    },
    {
      "id": "chess-puzzles",
      "rankingName": "Chess Puzzles",
      "rankingType": "benchmark",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 76,
      "url": "https://www.aiiq.org/api/rankings/chess-puzzles"
    },
    {
      "id": "critpt",
      "rankingName": "CritPt",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 149,
      "url": "https://www.aiiq.org/api/rankings/critpt"
    },
    {
      "id": "enterprise-ops-gym-aa",
      "rankingName": "EnterpriseOps-Gym-AA",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 19,
      "url": "https://www.aiiq.org/api/rankings/enterprise-ops-gym-aa"
    },
    {
      "id": "facts-grounding",
      "rankingName": "FACTS Grounding",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 33,
      "url": "https://www.aiiq.org/api/rankings/facts-grounding"
    },
    {
      "id": "frontiermath-t13",
      "rankingName": "FrontierMath Tier 1-3",
      "rankingType": "benchmark",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 78,
      "url": "https://www.aiiq.org/api/rankings/frontiermath-t13"
    },
    {
      "id": "frontiermath-t4",
      "rankingName": "FrontierMath Tier 4",
      "rankingType": "benchmark",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 66,
      "url": "https://www.aiiq.org/api/rankings/frontiermath-t4"
    },
    {
      "id": "gdp-pdf",
      "rankingName": "GDP.pdf",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 96,
      "url": "https://www.aiiq.org/api/rankings/gdp-pdf"
    },
    {
      "id": "gdpval-aa-v2",
      "rankingName": "GDPval-AA v2.1 normalized score",
      "rankingType": "benchmark",
      "dimension": "knowledge-work",
      "direction": "higher_is_better",
      "modelCount": 114,
      "url": "https://www.aiiq.org/api/rankings/gdpval-aa-v2"
    },
    {
      "id": "gpqa-diamond",
      "rankingName": "GPQA Diamond",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 145,
      "url": "https://www.aiiq.org/api/rankings/gpqa-diamond"
    },
    {
      "id": "humanitys-last-exam",
      "rankingName": "Humanity's Last Exam",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 157,
      "url": "https://www.aiiq.org/api/rankings/humanitys-last-exam"
    },
    {
      "id": "ifbench",
      "rankingName": "IFBench",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 95,
      "url": "https://www.aiiq.org/api/rankings/ifbench"
    },
    {
      "id": "ioi-v2",
      "rankingName": "IOI V2",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 41,
      "url": "https://www.aiiq.org/api/rankings/ioi-v2"
    },
    {
      "id": "livecodebench",
      "rankingName": "LiveCodeBench",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 61,
      "url": "https://www.aiiq.org/api/rankings/livecodebench"
    },
    {
      "id": "matharena",
      "rankingName": "MathArena",
      "rankingType": "benchmark",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 57,
      "url": "https://www.aiiq.org/api/rankings/matharena"
    },
    {
      "id": "mcp-atlas",
      "rankingName": "MCP Atlas",
      "rankingType": "benchmark",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 32,
      "url": "https://www.aiiq.org/api/rankings/mcp-atlas"
    },
    {
      "id": "mirrorcode",
      "rankingName": "MirrorCode",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 9,
      "url": "https://www.aiiq.org/api/rankings/mirrorcode"
    },
    {
      "id": "mmlu-pro",
      "rankingName": "MMLU-Pro",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 45,
      "url": "https://www.aiiq.org/api/rankings/mmlu-pro"
    },
    {
      "id": "mmmu-pro",
      "rankingName": "MMMU-Pro",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 84,
      "url": "https://www.aiiq.org/api/rankings/mmmu-pro"
    },
    {
      "id": "mystery-game-puzzles",
      "rankingName": "Mystery Game Puzzles",
      "rankingType": "benchmark",
      "dimension": "abstract-reasoning",
      "direction": "higher_is_better",
      "modelCount": 55,
      "url": "https://www.aiiq.org/api/rankings/mystery-game-puzzles"
    },
    {
      "id": "osworld-verified",
      "rankingName": "OSWorld-Verified",
      "rankingType": "benchmark",
      "dimension": "computer-use",
      "direction": "higher_is_better",
      "modelCount": 22,
      "url": "https://www.aiiq.org/api/rankings/osworld-verified"
    },
    {
      "id": "programbench-almost-resolved",
      "rankingName": "ProgramBench",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 61,
      "url": "https://www.aiiq.org/api/rankings/programbench-almost-resolved"
    },
    {
      "id": "proofbench-v1-1",
      "rankingName": "ProofBench 1.1",
      "rankingType": "benchmark",
      "dimension": "mathematical-reasoning",
      "direction": "higher_is_better",
      "modelCount": 46,
      "url": "https://www.aiiq.org/api/rankings/proofbench-v1-1"
    },
    {
      "id": "scicode",
      "rankingName": "SciCode",
      "rankingType": "benchmark",
      "dimension": "academic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 157,
      "url": "https://www.aiiq.org/api/rankings/scicode"
    },
    {
      "id": "simpleqa-verified",
      "rankingName": "SimpleQA Verified",
      "rankingType": "benchmark",
      "dimension": "reliability",
      "direction": "higher_is_better",
      "modelCount": 75,
      "url": "https://www.aiiq.org/api/rankings/simpleqa-verified"
    },
    {
      "id": "swe-rebench",
      "rankingName": "SWE-rebench",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 25,
      "url": "https://www.aiiq.org/api/rankings/swe-rebench"
    },
    {
      "id": "terminal-bench-4-0",
      "rankingName": "Terminal-Bench 4.0",
      "rankingType": "benchmark",
      "dimension": "programmatic-reasoning",
      "direction": "higher_is_better",
      "modelCount": 95,
      "url": "https://www.aiiq.org/api/rankings/terminal-bench-4-0"
    }
  ]
}