{
  "snapshot_date": "2026-11-11",
  "article_date": "2026-11-12",
  "status": "Approved for publication November 12, 2026",
  "benchmark_provider": "Artificial Analysis",
  "intelligence_version": "5.0.1",
  "not_original_model_tests": true,
  "scenario": {
    "uncached_input_tokens": 20000,
    "billable_output_tokens": 2000,
    "currency": "USD",
    "excludes": [
      "tools",
      "retries",
      "cache writes",
      "tax",
      "promotional credits"
    ],
    "carbon": "Global gross standard rates, not November credits-back effective rates",
    "haiku": "Rate tier <=100k input",
    "ember": "Rate tier <=200k input",
    "grok": "Base tier only"
  },
  "context_note": "Conservative 1,000,000-token bucket for million-token models, 500,000 for Grok; capacity is not comprehension",
  "weights": {
    "balanced": [
      0.5,
      0.2,
      0.25,
      0.05
    ],
    "quality": [
      0.7,
      0.2,
      0.05,
      0.05
    ],
    "budget": [
      0.3,
      0.2,
      0.45,
      0.05
    ]
  },
  "formula": "C=.6*Frontier+.4*Terminal; P=100*clip(log10(1/R)/3,0,1); W=100*clip(log2(K/32000)/5,0,1); weighted sum C,I,P,W",
  "overlap": "Intelligence Index already includes 25% coding; deliberate additional coding emphasis, not independent factors",
  "models": [
    {
      "name": "Claude Sonnet 5.5 (Max, Default Fallback)",
      "source": "https://artificialanalysis.ai/models/comparisons/claude-sonnet-5-5-vs-gpt-6-astra-xhigh",
      "intelligence": 55.0,
      "frontier": 41.7,
      "terminal": 63.6,
      "input": 2.0,
      "output": 10.0,
      "context": 1000000,
      "time": 32.7,
      "cost_task": 10.02,
      "label": "Claude Sonnet 5.5",
      "effort": "max",
      "coding": 50.46,
      "request_cost": 0.06,
      "price_utility": 40.72829165387854,
      "context_utility": 99.31568569324173,
      "score": 51.37785719813173,
      "quality_score": 53.32419886735601,
      "budget_score": 49.43151552890743,
      "score_rank": 1,
      "quality_score_rank": 4,
      "budget_score_rank": 3,
      "vendor_price_source": "https://platform.claude.com/docs/en/about-claude/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "55",
        "AA-Briefcase v1.1": "1816",
        "AutomationBench-AA": "71.8%",
        "Terminal-Bench 4.0": "63.6%",
        "FrontierCode-AA 2.0": "41.7%",
        "AA-Omniscience v1.1": "30",
        "GDP.pdf": "25.8%",
        "HLE-Diamond": "53.3%",
        "CritPt v2 (Hard Subset)": "59.0%",
        "Terminal-Bench-Science 0.1": "53.3%",
        "Cost": "",
        "Price per 1M Tokens": "$1.47",
        "Input Price per 1M Tokens": "$2.00",
        "Output Price per 1M Tokens": "$10.00",
        "Cache Hit Price per 1M Tokens": "$0.10",
        "Cost per Task": "$10.02",
        "Cost to Run Intelligence Index": "$7,981",
        "Token Use": "",
        "Output Tokens per Task": "285k",
        "Reasoning Tokens per Task": "206k",
        "Output Tokens to Run Intelligence Index": "335M",
        "Performance": "",
        "Time per Task": "32.7 min",
        "Output Speed": "145 tok/s",
        "Time to First Answer Token": "255.96s",
        "Time to First Token": "255.96s",
        "End-to-End Time (Mean of Single-Request Workloads)": "499.83s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "September 2026",
        "Knowledge Cutoff": "",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Claude Opus 5.5 (Max, Default Fallback)",
      "source": "https://artificialanalysis.ai/models/comparisons/claude-opus-5-5-vs-gpt-6-astra-vs-grok-4-5",
      "intelligence": 59.0,
      "frontier": 46.4,
      "terminal": 59.6,
      "input": 4.0,
      "output": 20.0,
      "context": 1000000,
      "time": 25.0,
      "cost_task": 10.04,
      "label": "Claude Opus 5.5",
      "effort": "max",
      "coding": 51.68000000000001,
      "request_cost": 0.12,
      "price_utility": 30.693958465079174,
      "context_utility": 99.31568569324173,
      "score": 50.27927390093188,
      "quality_score": 54.47648220791604,
      "budget_score": 46.08206559394772,
      "score_rank": 2,
      "quality_score_rank": 2,
      "budget_score_rank": 5,
      "vendor_price_source": "https://platform.claude.com/docs/en/about-claude/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "59",
        "AA-Briefcase v1.1": "1801",
        "AutomationBench-AA": "69.5%",
        "Terminal-Bench 4.0": "59.6%",
        "FrontierCode-AA 2.0": "46.4%",
        "AA-Omniscience v1.1": "48",
        "GDP.pdf": "26.2%",
        "HLE-Diamond": "67.9%",
        "CritPt v2 (Hard Subset)": "68.0%",
        "Terminal-Bench-Science 0.1": "59.0%",
        "Cost": "",
        "Price per 1M Tokens": "$2.94",
        "Input Price per 1M Tokens": "$4.00",
        "Output Price per 1M Tokens": "$20.00",
        "Cache Hit Price per 1M Tokens": "$0.20",
        "Cost per Task": "$10.04",
        "Cost to Run Intelligence Index": "$8,910",
        "Token Use": "",
        "Output Tokens per Task": "173k",
        "Reasoning Tokens per Task": "120k",
        "Output Tokens to Run Intelligence Index": "218M",
        "Performance": "",
        "Time per Task": "25.0 min",
        "Output Speed": "115 tok/s",
        "Time to First Answer Token": "353.75s",
        "Time to First Token": "353.75s",
        "End-to-End Time (Mean of Single-Request Workloads)": "725.27s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "September 2026",
        "Knowledge Cutoff": "",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Claude Haiku 5.5 (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/muse-ember-vs-claude-haiku-5-5",
      "intelligence": 36.0,
      "frontier": 34.9,
      "terminal": 32.8,
      "input": 0.1,
      "output": 0.5,
      "context": 1000000,
      "time": 14.8,
      "cost_task": 0.41,
      "label": "Claude Haiku 5.5",
      "effort": "max",
      "coding": 34.059999999999995,
      "request_cost": 0.003,
      "price_utility": 84.09595817601124,
      "context_utility": 99.31568569324173,
      "score": 50.21977382866489,
      "quality_score": 40.212582193462644,
      "budget_score": 60.226965463867145,
      "score_rank": 3,
      "quality_score_rank": 9,
      "budget_score_rank": 1,
      "vendor_price_source": "https://platform.claude.com/docs/en/about-claude/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "36",
        "AA-Briefcase v1.1": "1573",
        "AutomationBench-AA": "35.4%",
        "Terminal-Bench 4.0": "32.8%",
        "FrontierCode-AA 2.0": "34.9%",
        "AA-Omniscience v1.1": "12",
        "GDP.pdf": "20.8%",
        "HLE-Diamond": "29.6%",
        "CritPt v2 (Hard Subset)": "43.5%",
        "Terminal-Bench-Science 0.1": "20.0%",
        "Cost": "",
        "Price per 1M Tokens": "$0.077",
        "Input Price per 1M Tokens": "$0.10",
        "Output Price per 1M Tokens": "$0.50",
        "Cache Hit Price per 1M Tokens": "$0.01",
        "Cost per Task": "$0.41",
        "Cost to Run Intelligence Index": "$339",
        "Token Use": "",
        "Output Tokens per Task": "226k",
        "Reasoning Tokens per Task": "173k",
        "Output Tokens to Run Intelligence Index": "298M",
        "Performance": "",
        "Time per Task": "14.8 min",
        "Output Speed": "255 tok/s",
        "Time to First Answer Token": "185.68s",
        "Time to First Token": "185.68s",
        "End-to-End Time (Mean of Single-Request Workloads)": "336.27s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "October 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Gemini 4 Carbon (High)",
      "source": "https://artificialanalysis.ai/models/comparisons/gemini-4-carbon-vs-gpt-6-1-sol",
      "intelligence": 60.0,
      "frontier": 40.6,
      "terminal": 65.2,
      "input": 4.0,
      "output": 20.0,
      "context": 1000000,
      "time": 26.1,
      "cost_task": 7.76,
      "label": "Gemini 4 Carbon",
      "effort": "high",
      "coding": 50.44,
      "request_cost": 0.12,
      "price_utility": 30.693958465079174,
      "context_utility": 99.31568569324173,
      "score": 49.859273900931875,
      "quality_score": 53.808482207916036,
      "budget_score": 45.910065593947714,
      "score_rank": 4,
      "quality_score_rank": 3,
      "budget_score_rank": 6,
      "vendor_price_source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "60",
        "AA-Briefcase v1.1": "1611",
        "AutomationBench-AA": "77.8%",
        "Terminal-Bench 4.0": "65.2%",
        "FrontierCode-AA 2.0": "40.6%",
        "AA-Omniscience v1.1": "54",
        "GDP.pdf": "28.4%",
        "HLE-Diamond": "69.1%",
        "CritPt v2 (Hard Subset)": "69.0%",
        "Terminal-Bench-Science 0.1": "64.8%",
        "Cost": "",
        "Price per 1M Tokens": "$2.94",
        "Input Price per 1M Tokens": "$4.00",
        "Output Price per 1M Tokens": "$20.00",
        "Cache Hit Price per 1M Tokens": "$0.20",
        "Cost per Task": "$7.76",
        "Cost to Run Intelligence Index": "$8,240",
        "Token Use": "",
        "Output Tokens per Task": "123k",
        "Reasoning Tokens per Task": "72.2k",
        "Output Tokens to Run Intelligence Index": "205M",
        "Performance": "",
        "Time per Task": "26.1 min",
        "Output Speed": "78 tok/s",
        "Time to First Answer Token": "66.01s",
        "Time to First Token": "66.01s",
        "End-to-End Time (Mean of Single-Request Workloads)": "157.01s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "October 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text, image, speech, and video",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Claude Fable 5.2 (Max, Default Fallback)",
      "source": "https://artificialanalysis.ai/models/comparisons/claude-fable-5-2-vs-gpt-6-1-sol",
      "intelligence": 65.0,
      "frontier": 48.8,
      "terminal": 59.6,
      "input": 8.0,
      "output": 40.0,
      "context": 1000000,
      "time": 45.0,
      "cost_task": 26.54,
      "label": "Claude Fable 5.2",
      "effort": "max",
      "coding": 53.120000000000005,
      "request_cost": 0.24,
      "price_utility": 20.6596252762798,
      "context_utility": 99.31568569324173,
      "score": 49.69069060373204,
      "quality_score": 56.18276554847607,
      "budget_score": 43.198615658988,
      "score_rank": 5,
      "quality_score_rank": 1,
      "budget_score_rank": 8,
      "vendor_price_source": "https://platform.claude.com/docs/en/about-claude/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "65",
        "AA-Briefcase v1.1": "1785",
        "AutomationBench-AA": "76.4%",
        "Terminal-Bench 4.0": "59.6%",
        "FrontierCode-AA 2.0": "48.8%",
        "AA-Omniscience v1.1": "67",
        "GDP.pdf": "24.2%",
        "HLE-Diamond": "80.2%",
        "CritPt v2 (Hard Subset)": "76.5%",
        "Terminal-Bench-Science 0.1": "73.3%",
        "Cost": "",
        "Price per 1M Tokens": "$5.74",
        "Input Price per 1M Tokens": "$8.00",
        "Output Price per 1M Tokens": "$40.00",
        "Cache Hit Price per 1M Tokens": "$0.20",
        "Cost per Task": "$26.54",
        "Cost to Run Intelligence Index": "$22,191",
        "Token Use": "",
        "Output Tokens per Task": "274k",
        "Reasoning Tokens per Task": "193k",
        "Output Tokens to Run Intelligence Index": "275M",
        "Performance": "",
        "Time per Task": "45.0 min",
        "Output Speed": "101 tok/s",
        "Time to First Answer Token": "247.85s",
        "Time to First Token": "247.85s",
        "End-to-End Time (Mean of Single-Request Workloads)": "435.62s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "November 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "GPT-6.1 Sol (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/gpt-6-1-sol-vs-grok-4-7",
      "intelligence": 55.0,
      "frontier": 40.6,
      "terminal": 56.1,
      "input": 2.0,
      "output": 10.0,
      "context": 1000000,
      "time": 20.4,
      "cost_task": 1.27,
      "label": "GPT-6.1 Sol",
      "effort": "max",
      "coding": 46.8,
      "request_cost": 0.06,
      "price_utility": 40.72829165387854,
      "context_utility": 99.31568569324173,
      "score": 49.547857198131716,
      "quality_score": 50.76219886735601,
      "budget_score": 48.333515528907434,
      "score_rank": 6,
      "quality_score_rank": 6,
      "budget_score_rank": 4,
      "vendor_price_source": "https://developers.openai.com/api/docs/models/gpt-6.1-sol",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "55",
        "AA-Briefcase v1.1": "1555",
        "AutomationBench-AA": "64.9%",
        "Terminal-Bench 4.0": "56.1%",
        "FrontierCode-AA 2.0": "40.6%",
        "AA-Omniscience v1.1": "42",
        "GDP.pdf": "31.0%",
        "HLE-Diamond": "58.8%",
        "CritPt v2 (Hard Subset)": "69.5%",
        "Terminal-Bench-Science 0.1": "58.1%",
        "Cost": "",
        "Price per 1M Tokens": "$1.47",
        "Input Price per 1M Tokens": "$2.00",
        "Output Price per 1M Tokens": "$10.00",
        "Cache Hit Price per 1M Tokens": "$0.10",
        "Cost per Task": "$1.27",
        "Cost to Run Intelligence Index": "$1,124",
        "Token Use": "",
        "Output Tokens per Task": "67.8k",
        "Reasoning Tokens per Task": "49.2k",
        "Output Tokens to Run Intelligence Index": "63.7M",
        "Performance": "",
        "Time per Task": "20.4 min",
        "Output Speed": "55 tok/s",
        "Time to First Answer Token": "156.46s",
        "Time to First Token": "156.46s",
        "End-to-End Time (Mean of Single-Request Workloads)": "375.29s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "September 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Muse Ember (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/muse-ember-vs-claude-haiku-5-5",
      "intelligence": 36.0,
      "frontier": 27.2,
      "terminal": 40.9,
      "input": 0.1,
      "output": 0.5,
      "context": 1000000,
      "time": 3.1,
      "cost_task": 0.14,
      "label": "Muse Ember",
      "effort": "max",
      "coding": 32.68,
      "request_cost": 0.003,
      "price_utility": 84.09595817601124,
      "context_utility": 99.31568569324173,
      "score": 49.529773828664894,
      "quality_score": 39.24658219346264,
      "budget_score": 59.81296546386715,
      "score_rank": 7,
      "quality_score_rank": 10,
      "budget_score_rank": 2,
      "vendor_price_source": "https://dev.meta.ai/docs/pricing-rate-limits",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "36",
        "AA-Briefcase v1.1": "1521",
        "AutomationBench-AA": "68.1%",
        "Terminal-Bench 4.0": "40.9%",
        "FrontierCode-AA 2.0": "27.2%",
        "AA-Omniscience v1.1": "7",
        "GDP.pdf": "22.8%",
        "HLE-Diamond": "25.1%",
        "CritPt v2 (Hard Subset)": "39.0%",
        "Terminal-Bench-Science 0.1": "20.0%",
        "Cost": "",
        "Price per 1M Tokens": "$0.0728",
        "Input Price per 1M Tokens": "$0.10",
        "Output Price per 1M Tokens": "$0.50",
        "Cache Hit Price per 1M Tokens": "$0.004",
        "Cost per Task": "$0.14",
        "Cost to Run Intelligence Index": "$117",
        "Token Use": "",
        "Output Tokens per Task": "84.5k",
        "Reasoning Tokens per Task": "52.2k",
        "Output Tokens to Run Intelligence Index": "115M",
        "Performance": "",
        "Time per Task": "3.1 min",
        "Output Speed": "454 tok/s",
        "Time to First Answer Token": "4.96s",
        "Time to First Token": "4.96s",
        "End-to-End Time (Mean of Single-Request Workloads)": "13.26s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,573 A4 pages of size 12 Arial font",
        "Release Date": "November 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Kimi K3.1 Pro",
      "source": "https://artificialanalysis.ai/models/comparisons/kimi-k3-1-pro-vs-glm-5-3",
      "intelligence": 46.0,
      "frontier": 25.5,
      "terminal": 67.2,
      "input": 2.0,
      "output": 10.0,
      "context": 1000000,
      "time": 37.3,
      "cost_task": 3.68,
      "label": "Kimi K3.1 Pro",
      "effort": "max",
      "coding": 42.18,
      "request_cost": 0.06,
      "price_utility": 40.72829165387854,
      "context_utility": 99.31568569324173,
      "score": 45.43785719813172,
      "quality_score": 45.728198867356014,
      "budget_score": 45.14751552890743,
      "score_rank": 8,
      "quality_score_rank": 7,
      "budget_score_rank": 7,
      "vendor_price_source": "https://www.kimi.ai/resources/kimi-k3-1-pro-pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "46",
        "AA-Briefcase v1.1": "1674",
        "AutomationBench-AA": "73.3%",
        "Terminal-Bench 4.0": "67.2%",
        "FrontierCode-AA 2.0": "25.5%",
        "AA-Omniscience v1.1": "30",
        "GDP.pdf": "31.4%",
        "HLE-Diamond": "37.3%",
        "CritPt v2 (Hard Subset)": "49.5%",
        "Terminal-Bench-Science 0.1": "34.3%",
        "Cost": "",
        "Price per 1M Tokens": "$1.54",
        "Input Price per 1M Tokens": "$2.00",
        "Output Price per 1M Tokens": "$10.00",
        "Cache Hit Price per 1M Tokens": "$0.20",
        "Cost per Task": "$3.68",
        "Cost to Run Intelligence Index": "$3,153",
        "Token Use": "",
        "Output Tokens per Task": "72.2k",
        "Reasoning Tokens per Task": "44.2k",
        "Output Tokens to Run Intelligence Index": "113M",
        "Performance": "",
        "Time per Task": "37.3 min",
        "Output Speed": "32 tok/s",
        "Time to First Answer Token": "169.69s",
        "Time to First Token": "3.97s",
        "End-to-End Time (Mean of Single-Request Workloads)": "416.84s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "October 2026",
        "Total parameters": "2.8T",
        "Active parameters": "104B",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "Yes",
        "License": "Kimi K3.1-Pro Licence",
        "License Supports Commercial Use Without Restrictions": "Yes"
      },
      "effort_provenance": "Vendor default max; AA comparison row does not explicitly suffix effort. Not a recovered per-run configuration."
    },
    {
      "name": "GPT-6 Astra (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/claude-opus-5-5-vs-gpt-6-astra-vs-grok-4-5",
      "intelligence": 57.0,
      "frontier": 42.4,
      "terminal": 59.1,
      "input": 10.0,
      "output": 50.0,
      "context": 1000000,
      "time": 23.0,
      "cost_task": 6.24,
      "label": "GPT-6 Astra",
      "effort": "max",
      "coding": 49.08,
      "request_cost": 0.30000000000000004,
      "price_utility": 17.429291509344583,
      "context_utility": 99.31568569324173,
      "score": 45.26310716199823,
      "quality_score": 51.59324886012931,
      "budget_score": 38.93296546386715,
      "score_rank": 9,
      "quality_score_rank": 5,
      "budget_score_rank": 10,
      "vendor_price_source": "https://developers.openai.com/api/docs/models/gpt-6-astra",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "57",
        "AA-Briefcase v1.1": "1569",
        "AutomationBench-AA": "68.5%",
        "Terminal-Bench 4.0": "59.1%",
        "FrontierCode-AA 2.0": "42.4%",
        "AA-Omniscience v1.1": "42",
        "GDP.pdf": "31.0%",
        "HLE-Diamond": "63.8%",
        "CritPt v2 (Hard Subset)": "70.5%",
        "Terminal-Bench-Science 0.1": "63.3%",
        "Cost": "",
        "Price per 1M Tokens": "$7.70",
        "Input Price per 1M Tokens": "$10.00",
        "Output Price per 1M Tokens": "$50.00",
        "Cache Hit Price per 1M Tokens": "$1.00",
        "Cost per Task": "$6.24",
        "Cost to Run Intelligence Index": "$5,661",
        "Token Use": "",
        "Output Tokens per Task": "52k",
        "Reasoning Tokens per Task": "35.8k",
        "Output Tokens to Run Intelligence Index": "54.3M",
        "Performance": "",
        "Time per Task": "23.0 min",
        "Output Speed": "38 tok/s",
        "Time to First Answer Token": "249.02s",
        "Time to First Token": "249.02s",
        "End-to-End Time (Mean of Single-Request Workloads)": "581.42s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "September 2026",
        "Knowledge Cutoff": "April 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Muse Wonder (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/muse-wonder-vs-glm-5-3",
      "intelligence": 44.0,
      "frontier": 27.9,
      "terminal": 52.0,
      "input": 4.0,
      "output": 20.0,
      "context": 1000000,
      "time": 13.3,
      "cost_task": 6.95,
      "label": "Muse Wonder",
      "effort": "max",
      "coding": 37.54,
      "request_cost": 0.12,
      "price_utility": 30.693958465079174,
      "context_utility": 99.31568569324173,
      "score": 40.209273900931876,
      "quality_score": 41.578482207916046,
      "budget_score": 38.84006559394771,
      "score_rank": 10,
      "quality_score_rank": 8,
      "budget_score_rank": 11,
      "vendor_price_source": "https://dev.meta.ai/docs/pricing-rate-limits",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "44",
        "AA-Briefcase v1.1": "1659",
        "AutomationBench-AA": "68.1%",
        "Terminal-Bench 4.0": "52.0%",
        "FrontierCode-AA 2.0": "27.9%",
        "AA-Omniscience v1.1": "25",
        "GDP.pdf": "30.6%",
        "HLE-Diamond": "39.0%",
        "CritPt v2 (Hard Subset)": "50.5%",
        "Terminal-Bench-Science 0.1": "25.7%",
        "Cost": "",
        "Price per 1M Tokens": "$2.94",
        "Input Price per 1M Tokens": "$4.00",
        "Output Price per 1M Tokens": "$20.00",
        "Cache Hit Price per 1M Tokens": "$0.20",
        "Cost per Task": "$6.95",
        "Cost to Run Intelligence Index": "$5,414",
        "Token Use": "",
        "Output Tokens per Task": "96.1k",
        "Reasoning Tokens per Task": "58.6k",
        "Output Tokens to Run Intelligence Index": "113M",
        "Performance": "",
        "Time per Task": "13.3 min",
        "Output Speed": "120 tok/s",
        "Time to First Answer Token": "22.49s",
        "Time to First Token": "22.49s",
        "End-to-End Time (Mean of Single-Request Workloads)": "60.75s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,573 A4 pages of size 12 Arial font",
        "Release Date": "October 2026",
        "Total parameters": "",
        "Active parameters": "",
        "Reasoning": "Yes",
        "Input modality": "Supports: text, image, speech, and video",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No",
        "License": "",
        "License Supports Commercial Use Without Restrictions": ""
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "GLM-5.3 (Max)",
      "source": "https://artificialanalysis.ai/models/comparisons/muse-wonder-vs-glm-5-3",
      "intelligence": 32.0,
      "frontier": 21.6,
      "terminal": 41.9,
      "input": 1.4,
      "output": 4.4,
      "context": 1000000,
      "time": 28.3,
      "cost_task": 5.01,
      "label": "GLM-5.3",
      "effort": "max",
      "coding": 29.720000000000002,
      "request_cost": 0.0368,
      "price_utility": 47.80507271088275,
      "context_utility": 99.31568569324173,
      "score": 38.177052462382775,
      "quality_score": 34.56003792020623,
      "budget_score": 41.794067004559324,
      "score_rank": 11,
      "quality_score_rank": 11,
      "budget_score_rank": 9,
      "vendor_price_source": "https://docs.z.ai/guides/overview/pricing",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "32",
        "AA-Briefcase v1.1": "1503",
        "AutomationBench-AA": "62.2%",
        "Terminal-Bench 4.0": "41.9%",
        "FrontierCode-AA 2.0": "21.6%",
        "AA-Omniscience v1.1": "6",
        "GDP.pdf": "11.2%",
        "HLE-Diamond": "17.9%",
        "CritPt v2 (Hard Subset)": "33.5%",
        "Terminal-Bench-Science 0.1": "9.5%",
        "Cost": "",
        "Price per 1M Tokens": "$0.90",
        "Input Price per 1M Tokens": "$1.40",
        "Output Price per 1M Tokens": "$4.40",
        "Cache Hit Price per 1M Tokens": "$0.26",
        "Cost per Task": "$5.01",
        "Cost to Run Intelligence Index": "$3,496",
        "Token Use": "",
        "Output Tokens per Task": "119k",
        "Reasoning Tokens per Task": "84.8k",
        "Output Tokens to Run Intelligence Index": "135M",
        "Performance": "",
        "Time per Task": "28.3 min",
        "Output Speed": "70 tok/s",
        "Time to First Answer Token": "59.56s",
        "Time to First Token": "3.31s",
        "End-to-End Time (Mean of Single-Request Workloads)": "128.39s",
        "Technical specifications": "",
        "Context Window": "1M tokens~1,500 A4 pages of size 12 Arial font",
        "Release Date": "August 2026",
        "Total parameters": "753B",
        "Active parameters": "40B",
        "Reasoning": "Yes",
        "Input modality": "Supports: text",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "Yes",
        "License": "GLM-5.3 License",
        "License Supports Commercial Use Without Restrictions": "Yes"
      },
      "effort_provenance": "Explicit AA row label"
    },
    {
      "name": "Grok 4.7 (Xhigh)",
      "source": "https://artificialanalysis.ai/models/comparisons/gpt-6-1-sol-vs-grok-4-7",
      "intelligence": 35.0,
      "frontier": 28.7,
      "terminal": 25.8,
      "input": 2.0,
      "output": 6.0,
      "context": 500000,
      "time": 27.4,
      "cost_task": 7.77,
      "label": "Grok 4.7",
      "effort": "xhigh",
      "coding": 27.54,
      "request_cost": 0.052000000000000005,
      "price_utility": 42.79988854550669,
      "context_utility": 79.31568569324175,
      "score": 35.43575642103876,
      "quality_score": 32.38377871193742,
      "budget_score": 38.4877341301401,
      "score_rank": 12,
      "quality_score_rank": 12,
      "budget_score_rank": 12,
      "vendor_price_source": "https://docs.x.ai/developers/models/grok-4.7",
      "raw_source_fields": {
        "Intelligence": "",
        "Intelligence Index": "35",
        "AA-Briefcase v1.1": "1636",
        "AutomationBench-AA": "65.6%",
        "Terminal-Bench 4.0": "25.8%",
        "FrontierCode-AA 2.0": "28.7%",
        "AA-Omniscience v1.1": "31",
        "GDP.pdf": "20.0%",
        "HLE-Diamond": "23.2%",
        "CritPt v2 (Hard Subset)": "33.5%",
        "Terminal-Bench-Science 0.1": "11.4%",
        "Cost": "",
        "Price per 1M Tokens": "$1.35",
        "Input Price per 1M Tokens": "$2.00",
        "Output Price per 1M Tokens": "$6.00",
        "Cache Hit Price per 1M Tokens": "$0.50",
        "Cost per Task": "$7.77",
        "Cost to Run Intelligence Index": "$5,905",
        "Token Use": "",
        "Output Tokens per Task": "105k",
        "Reasoning Tokens per Task": "69.4k",
        "Output Tokens to Run Intelligence Index": "148M",
        "Performance": "",
        "Time per Task": "27.4 min",
        "Output Speed": "63 tok/s",
        "Time to First Answer Token": "57.56s",
        "Time to First Token": "57.56s",
        "End-to-End Time (Mean of Single-Request Workloads)": "129.96s",
        "Technical specifications": "",
        "Context Window": "500k tokens~750 A4 pages of size 12 Arial font",
        "Release Date": "September 2026",
        "Reasoning": "Yes",
        "Input modality": "Supports: text and image",
        "Output modality": "Supports: text",
        "Open Source (Weights)": "No"
      },
      "effort_provenance": "Explicit AA row label"
    }
  ],
  "sources": [
    {
      "number": 1,
      "title": "Artificial Analysis Intelligence Index methodology",
      "url": "https://artificialanalysis.ai/methodology/intelligence-benchmarking"
    },
    {
      "number": 2,
      "title": "FrontierCode-AA 2.0 methodology",
      "url": "https://artificialanalysis.ai/evaluations/frontiercode-2-0"
    },
    {
      "number": 3,
      "title": "Fable 5.2 and Sol comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/claude-fable-5-2-vs-gpt-6-1-sol"
    },
    {
      "number": 4,
      "title": "Carbon and Sol comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/gemini-4-carbon-vs-gpt-6-1-sol"
    },
    {
      "number": 5,
      "title": "Opus 5.5 and Astra comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/claude-opus-5-5-vs-gpt-6-astra-vs-grok-4-5"
    },
    {
      "number": 6,
      "title": "Sonnet 5.5 comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/claude-sonnet-5-5-vs-gpt-6-astra-xhigh"
    },
    {
      "number": 7,
      "title": "Kimi K3.1 Pro and GLM-5.3 comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/kimi-k3-1-pro-vs-glm-5-3"
    },
    {
      "number": 8,
      "title": "Muse Wonder comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/muse-wonder-vs-glm-5-3"
    },
    {
      "number": 9,
      "title": "Muse Ember and Haiku 5.5 comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/muse-ember-vs-claude-haiku-5-5"
    },
    {
      "number": 10,
      "title": "Sol and Grok 4.7 comparison",
      "url": "https://artificialanalysis.ai/models/comparisons/gpt-6-1-sol-vs-grok-4-7"
    },
    {
      "number": 11,
      "title": "Anthropic API pricing",
      "url": "https://platform.claude.com/docs/en/about-claude/pricing"
    },
    {
      "number": 12,
      "title": "OpenAI Sol model documentation",
      "url": "https://developers.openai.com/api/docs/models/gpt-6.1-sol"
    },
    {
      "number": 13,
      "title": "OpenAI Astra model documentation",
      "url": "https://developers.openai.com/api/docs/models/gpt-6-astra"
    },
    {
      "number": 14,
      "title": "Google Cloud Agent Platform pricing",
      "url": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing"
    },
    {
      "number": 15,
      "title": "Kimi K3.1 Pro pricing",
      "url": "https://www.kimi.ai/resources/kimi-k3-1-pro-pricing"
    },
    {
      "number": 16,
      "title": "Z.ai API pricing",
      "url": "https://docs.z.ai/guides/overview/pricing"
    },
    {
      "number": 17,
      "title": "Meta Model API pricing and rate limits",
      "url": "https://dev.meta.ai/docs/pricing-rate-limits"
    },
    {
      "number": 18,
      "title": "Grok 4.7 model documentation",
      "url": "https://docs.x.ai/developers/models/grok-4.7"
    },
    {
      "number": 19,
      "title": "Google Carbon model documentation",
      "url": "https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-carbon"
    },
    {
      "number": 20,
      "title": "Muse Wonder API documentation",
      "url": "https://dev.meta.ai/docs/muse-wonder"
    },
    {
      "number": 21,
      "title": "Claude model overview",
      "url": "https://platform.claude.com/docs/en/models/overview"
    },
    {
      "number": 22,
      "title": "Muse Ember model documentation",
      "url": "https://dev.meta.ai/models/muse-ember"
    }
  ],
  "rechecked_date": "2026-11-12"
}