{
 "Long Agentic": {
  "parts": [
   {
    "name": "Finance & Compliance Apps",
    "gloss": "Payroll, budgets, tax claims and audit records: keep the money and the paper trail right as the rules change, turn after turn.",
    "tasks": 3
   },
   {
    "name": "Operations & Commerce Apps",
    "gloss": "Ticketing, equipment hire, stock and site records: keep a live operational app correct as the business changes around it.",
    "tasks": 4
   },
   {
    "name": "Long-Run Operations",
    "gloss": "Run an existing system through simulated months of releases, closes and change requests, where early choices compound.",
    "tasks": 3
   },
   {
    "name": "Delegated Operations",
    "gloss": "Keep a small business's books and duties right for weeks while its owner is away, through a command someone else follows to the letter; what matters shows only in the data.",
    "tasks": 4
   },
   {
    "name": "Systems Engineering",
    "gloss": "Build and repair complex systems from a specification, then keep them running under realistic load and failures.",
    "tasks": 2
   },
   {
    "name": "Performance Optimisation",
    "gloss": "Make real programs faster and real schedules tighter, scored on measured results rather than judgement.",
    "tasks": 2
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Finance & Compliance Apps": 94.4,
     "Operations & Commerce Apps": 77.1,
     "Long-Run Operations": 39.0,
     "Delegated Operations": 72.6,
     "Systems Engineering": 93.2,
     "Performance Optimisation": 74.8
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Finance & Compliance Apps": 100.0,
     "Operations & Commerce Apps": 89.6,
     "Long-Run Operations": 59.2,
     "Delegated Operations": 73.8,
     "Systems Engineering": 97.0,
     "Performance Optimisation": 83.2
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Finance & Compliance Apps": 44.4,
     "Operations & Commerce Apps": 83.3,
     "Long-Run Operations": 25.9,
     "Delegated Operations": 6.8,
     "Systems Engineering": 85.7,
     "Performance Optimisation": 60.1
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Finance & Compliance Apps": 88.9,
     "Operations & Commerce Apps": 95.8,
     "Long-Run Operations": 58.6,
     "Delegated Operations": 33.0,
     "Systems Engineering": 97.0,
     "Performance Optimisation": 79.3
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Finance & Compliance Apps": 88.9,
     "Operations & Commerce Apps": 89.6,
     "Long-Run Operations": 43.4,
     "Delegated Operations": 13.8,
     "Systems Engineering": 96.7,
     "Performance Optimisation": 68.7
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Finance & Compliance Apps": 11.1,
     "Operations & Commerce Apps": 50.0,
     "Long-Run Operations": 32.5,
     "Delegated Operations": 32.6,
     "Systems Engineering": 73.8,
     "Performance Optimisation": 44.4
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Finance & Compliance Apps": 5.6,
     "Operations & Commerce Apps": 12.5,
     "Long-Run Operations": 5.8,
     "Delegated Operations": 0.7,
     "Systems Engineering": 65.1,
     "Performance Optimisation": 39.6
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Finance & Compliance Apps": 88.9,
     "Operations & Commerce Apps": 50.0,
     "Long-Run Operations": 38.2,
     "Delegated Operations": 47.4,
     "Systems Engineering": 95.5,
     "Performance Optimisation": 45.6
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Finance & Compliance Apps": 88.9,
     "Operations & Commerce Apps": 100.0,
     "Long-Run Operations": 60.0,
     "Delegated Operations": 45.1,
     "Systems Engineering": 95.8,
     "Performance Optimisation": 45.0
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 0.0,
     "Long-Run Operations": 0.0,
     "Delegated Operations": 5.2,
     "Systems Engineering": 27.8,
     "Performance Optimisation": 38.7
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Finance & Compliance Apps": 5.6,
     "Operations & Commerce Apps": 27.1,
     "Long-Run Operations": 16.9,
     "Delegated Operations": 8.7,
     "Systems Engineering": 39.4,
     "Performance Optimisation": 42.0
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Finance & Compliance Apps": 11.1,
     "Operations & Commerce Apps": 33.3,
     "Long-Run Operations": 20.7,
     "Delegated Operations": 2.3,
     "Systems Engineering": 73.1,
     "Performance Optimisation": 40.9
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Finance & Compliance Apps": 88.9,
     "Operations & Commerce Apps": 100.0,
     "Long-Run Operations": 26.1,
     "Delegated Operations": 71.9,
     "Systems Engineering": 97.3,
     "Performance Optimisation": 67.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Finance & Compliance Apps": 22.2,
     "Operations & Commerce Apps": 31.2,
     "Long-Run Operations": 16.9,
     "Delegated Operations": 1.2,
     "Systems Engineering": 63.2,
     "Performance Optimisation": 35.4
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Finance & Compliance Apps": 61.1,
     "Operations & Commerce Apps": 100.0,
     "Long-Run Operations": 27.5,
     "Delegated Operations": 50.3,
     "Systems Engineering": 97.2,
     "Performance Optimisation": 59.6
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Finance & Compliance Apps": 61.1,
     "Operations & Commerce Apps": 93.8,
     "Long-Run Operations": 34.8,
     "Delegated Operations": 1.5,
     "Systems Engineering": 96.9,
     "Performance Optimisation": 74.6
    }
   },
   "grok-4.6": {
    "scores": {
     "Finance & Compliance Apps": 16.7,
     "Operations & Commerce Apps": 8.3,
     "Long-Run Operations": 6.3,
     "Delegated Operations": 2.9,
     "Systems Engineering": 71.2,
     "Performance Optimisation": 40.6
    }
   },
   "grok-4.7": {
    "scores": {
     "Finance & Compliance Apps": 66.7,
     "Operations & Commerce Apps": 41.7,
     "Long-Run Operations": 7.1,
     "Delegated Operations": 49.4,
     "Systems Engineering": 70.9,
     "Performance Optimisation": 44.4
    }
   },
   "inkling": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 0.0,
     "Long-Run Operations": 1.8,
     "Delegated Operations": 4.9,
     "Systems Engineering": 38.6,
     "Performance Optimisation": 37.5
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Finance & Compliance Apps": 27.8,
     "Operations & Commerce Apps": 29.2,
     "Long-Run Operations": 31.2,
     "Delegated Operations": 33.7,
     "Systems Engineering": 73.5,
     "Performance Optimisation": 61.0
    }
   },
   "kimi-k3": {
    "scores": {
     "Finance & Compliance Apps": 38.9,
     "Operations & Commerce Apps": 37.5,
     "Long-Run Operations": 11.6,
     "Delegated Operations": 31.8,
     "Systems Engineering": 52.5,
     "Performance Optimisation": 57.4
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Finance & Compliance Apps": 66.7,
     "Operations & Commerce Apps": 39.6,
     "Long-Run Operations": 6.3,
     "Delegated Operations": 6.9,
     "Systems Engineering": 79.5,
     "Performance Optimisation": 45.6
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Finance & Compliance Apps": 22.2,
     "Operations & Commerce Apps": 29.2,
     "Long-Run Operations": 10.7,
     "Delegated Operations": 0.6,
     "Systems Engineering": 71.8,
     "Performance Optimisation": 50.2
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Finance & Compliance Apps": 27.8,
     "Operations & Commerce Apps": 20.8,
     "Long-Run Operations": 3.6,
     "Delegated Operations": 25.3,
     "Systems Engineering": 46.4,
     "Performance Optimisation": 36.4
    }
   },
   "mistral-large-4": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 0.0,
     "Long-Run Operations": 3.6,
     "Delegated Operations": 5.7,
     "Systems Engineering": 28.9,
     "Performance Optimisation": 37.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Finance & Compliance Apps": 55.6,
     "Operations & Commerce Apps": 70.8,
     "Long-Run Operations": 34.0,
     "Delegated Operations": 14.1,
     "Systems Engineering": 85.5,
     "Performance Optimisation": 41.2
    }
   },
   "muse-spark-open": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 4.2,
     "Long-Run Operations": 2.2,
     "Delegated Operations": 3.5,
     "Systems Engineering": 51.0,
     "Performance Optimisation": 39.7
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Finance & Compliance Apps": 50.0,
     "Operations & Commerce Apps": 60.4,
     "Long-Run Operations": 48.8,
     "Delegated Operations": 7.8,
     "Systems Engineering": 82.7,
     "Performance Optimisation": 49.0
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 0.0,
     "Long-Run Operations": 1.9,
     "Delegated Operations": 5.2,
     "Systems Engineering": 52.8,
     "Performance Optimisation": 32.7
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 14.6,
     "Long-Run Operations": 1.1,
     "Delegated Operations": 5.0,
     "Systems Engineering": 32.6,
     "Performance Optimisation": 34.5
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Finance & Compliance Apps": 5.6,
     "Operations & Commerce Apps": 4.2,
     "Long-Run Operations": 15.5,
     "Delegated Operations": 3.2,
     "Systems Engineering": 49.5,
     "Performance Optimisation": 47.1
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Finance & Compliance Apps": 11.1,
     "Operations & Commerce Apps": 45.8,
     "Long-Run Operations": 6.5,
     "Delegated Operations": 6.4,
     "Systems Engineering": 66.3,
     "Performance Optimisation": 41.9
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Finance & Compliance Apps": 0.0,
     "Operations & Commerce Apps": 0.0,
     "Long-Run Operations": 3.4,
     "Delegated Operations": 6.4,
     "Systems Engineering": 53.2,
     "Performance Optimisation": 30.4
    }
   },
   "step-5-preview": {
    "scores": {
     "Finance & Compliance Apps": 5.6,
     "Operations & Commerce Apps": 20.8,
     "Long-Run Operations": 9.9,
     "Delegated Operations": 1.4,
     "Systems Engineering": 63.5,
     "Performance Optimisation": 43.0
    }
   }
  }
 },
 "Reasoning": {
  "parts": [
   {
    "name": "Theory of mind",
    "gloss": "Track who knows or believes what in a short story.",
    "tasks": 52
   },
   {
    "name": "Zebra puzzles",
    "gloss": "Deduce who has which attribute from interlocking clues.",
    "tasks": 50
   },
   {
    "name": "Spatial",
    "gloss": "Reason about shapes and positions in 2D and 3D.",
    "tasks": 50
   },
   {
    "name": "Logic with navigation",
    "gloss": "Work out who tells the truth from a chain of statements.",
    "tasks": 50
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Theory of mind": 80.8,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 86.0
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Theory of mind": 73.1,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 82.0
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Theory of mind": 78.8,
     "Zebra puzzles": 96.8,
     "Spatial": 88.0,
     "Logic with navigation": 66.0
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Theory of mind": 84.6,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 86.0
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 82.0
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Theory of mind": 88.5,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 70.0
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Theory of mind": 76.9,
     "Zebra puzzles": 98.2,
     "Spatial": 100.0,
     "Logic with navigation": 82.0
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Theory of mind": 92.3,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 88.0
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 76.0
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Theory of mind": 73.1,
     "Zebra puzzles": 36.0,
     "Spatial": 96.0,
     "Logic with navigation": 68.0
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Theory of mind": 73.1,
     "Zebra puzzles": 79.5,
     "Spatial": 96.0,
     "Logic with navigation": 62.0
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 76.0
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Theory of mind": 84.6,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 88.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Theory of mind": 73.1,
     "Zebra puzzles": 96.0,
     "Spatial": 96.0,
     "Logic with navigation": 62.0
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 86.0
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Theory of mind": 84.6,
     "Zebra puzzles": 100.0,
     "Spatial": 98.0,
     "Logic with navigation": 72.0
    }
   },
   "grok-4.6": {
    "scores": {
     "Theory of mind": 90.4,
     "Zebra puzzles": 96.5,
     "Spatial": 98.0,
     "Logic with navigation": 80.0
    }
   },
   "grok-4.7": {
    "scores": {
     "Theory of mind": 84.6,
     "Zebra puzzles": 86.0,
     "Spatial": 94.0,
     "Logic with navigation": 66.0
    }
   },
   "inkling": {
    "scores": {
     "Theory of mind": 65.4,
     "Zebra puzzles": 88.0,
     "Spatial": 94.0,
     "Logic with navigation": 66.0
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 78.0
    }
   },
   "kimi-k3": {
    "scores": {
     "Theory of mind": 82.7,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 80.0
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 72.0,
     "Spatial": 94.0,
     "Logic with navigation": 66.0
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Theory of mind": 63.5,
     "Zebra puzzles": 100.0,
     "Spatial": 96.0,
     "Logic with navigation": 72.0
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Theory of mind": 88.5,
     "Zebra puzzles": 94.5,
     "Spatial": 94.0,
     "Logic with navigation": 64.0
    }
   },
   "mistral-large-4": {
    "scores": {
     "Theory of mind": 80.8,
     "Zebra puzzles": 93.0,
     "Spatial": 94.0,
     "Logic with navigation": 68.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Theory of mind": 84.6,
     "Zebra puzzles": 98.0,
     "Spatial": 100.0,
     "Logic with navigation": 76.0
    }
   },
   "muse-spark-open": {
    "scores": {
     "Theory of mind": 76.9,
     "Zebra puzzles": 40.0,
     "Spatial": 100.0,
     "Logic with navigation": 78.0
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Theory of mind": 86.5,
     "Zebra puzzles": 100.0,
     "Spatial": 96.0,
     "Logic with navigation": 80.0
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Theory of mind": 63.5,
     "Zebra puzzles": 63.2,
     "Spatial": 100.0,
     "Logic with navigation": 60.0
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Theory of mind": 75.0,
     "Zebra puzzles": 67.5,
     "Spatial": 98.0,
     "Logic with navigation": 58.0
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Theory of mind": 59.6,
     "Zebra puzzles": 96.5,
     "Spatial": 96.0,
     "Logic with navigation": 68.0
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Theory of mind": 78.8,
     "Zebra puzzles": 100.0,
     "Spatial": 100.0,
     "Logic with navigation": 74.0
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Theory of mind": 75.0,
     "Zebra puzzles": 67.5,
     "Spatial": 98.0,
     "Logic with navigation": 60.0
    }
   },
   "step-5-preview": {
    "scores": {
     "Theory of mind": 75.0,
     "Zebra puzzles": 90.0,
     "Spatial": 98.0,
     "Logic with navigation": 72.0
    }
   }
  }
 },
 "Coding": {
  "parts": [
   {
    "name": "LLM API Integration",
    "gloss": "Provider wire formats and request/response handling.",
    "tasks": 3
   },
   {
    "name": "Agent Loop",
    "gloss": "Agent runtime control flow.",
    "tasks": 3
   },
   {
    "name": "Billing & Limits",
    "gloss": "Metering, quotas and limit enforcement.",
    "tasks": 3
   },
   {
    "name": "Cloud Infrastructure",
    "gloss": "Storage, compute and cloud service plumbing.",
    "tasks": 3
   },
   {
    "name": "Product & UX",
    "gloss": "User-facing product behaviour.",
    "tasks": 4
   },
   {
    "name": "Memory & Tools",
    "gloss": "Memory handling and tool integrations.",
    "tasks": 3
   },
   {
    "name": "Hard Fixes",
    "gloss": "Multi-site product fixes whose hidden tests cover every brief-stated requirement; partial credit halves per broken one.",
    "tasks": 5
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 100.0,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 83.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 63.9
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 100.0,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 88.9,
     "Product & UX": 58.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 63.6
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 66.7,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 66.7,
     "Product & UX": 66.7,
     "Memory & Tools": 100.0,
     "Hard Fixes": 62.5
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 96.7,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 83.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 53.5
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 100.0,
     "Billing & Limits": 77.8,
     "Cloud Infrastructure": 96.7,
     "Product & UX": 75.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 26.0
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 66.7,
     "Billing & Limits": 66.7,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 91.7,
     "Memory & Tools": 100.0,
     "Hard Fixes": 62.7
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 66.7,
     "Billing & Limits": 66.7,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 58.3,
     "Memory & Tools": 66.7,
     "Hard Fixes": 38.7
    }
   },
   "gemini-4-argon": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 66.7,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 91.7,
     "Memory & Tools": 91.7,
     "Hard Fixes": 63.2
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 100.0,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 100.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 56.7
    }
   },
   "gemma-4-31b": {
    "scores": {
     "LLM API Integration": 38.9,
     "Agent Loop": 16.7,
     "Billing & Limits": 0.0,
     "Cloud Infrastructure": 27.8,
     "Product & UX": 0.0,
     "Memory & Tools": 57.1,
     "Hard Fixes": 13.3
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "LLM API Integration": 83.5,
     "Agent Loop": 66.7,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 83.3,
     "Product & UX": 33.3,
     "Memory & Tools": 95.2,
     "Hard Fixes": 59.1
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 100.0,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 75.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 54.3
    }
   },
   "gpt-6-astra": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 83.3,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 100.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 70.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 83.3,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 66.7,
     "Memory & Tools": 33.3,
     "Hard Fixes": 56.7
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "LLM API Integration": 77.8,
     "Agent Loop": 83.3,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 100.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 37.5
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "LLM API Integration": 77.8,
     "Agent Loop": 100.0,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 75.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 81.7
    }
   },
   "grok-4.6": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 66.7,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 75.0,
     "Memory & Tools": 100.0,
     "Hard Fixes": 51.3
    }
   },
   "grok-4.7": {
    "scores": {
     "LLM API Integration": 16.7,
     "Agent Loop": 100.0,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 83.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 34.0
    }
   },
   "inkling": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 33.3,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 66.7,
     "Product & UX": 8.3,
     "Memory & Tools": 57.1,
     "Hard Fixes": 36.3
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 66.7,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 88.9,
     "Product & UX": 33.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 48.9
    }
   },
   "kimi-k3": {
    "scores": {
     "LLM API Integration": 50.0,
     "Agent Loop": 0.0,
     "Billing & Limits": 66.7,
     "Cloud Infrastructure": 50.0,
     "Product & UX": 58.3,
     "Memory & Tools": 62.0,
     "Hard Fixes": 60.0
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "LLM API Integration": 33.3,
     "Agent Loop": 66.7,
     "Billing & Limits": 100.0,
     "Cloud Infrastructure": 83.3,
     "Product & UX": 58.3,
     "Memory & Tools": 63.3,
     "Hard Fixes": 55.1
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "LLM API Integration": 33.3,
     "Agent Loop": 100.0,
     "Billing & Limits": 66.7,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 83.3,
     "Memory & Tools": 58.6,
     "Hard Fixes": 48.3
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 100.0,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 58.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 47.8
    }
   },
   "mistral-large-4": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 33.3,
     "Billing & Limits": 61.1,
     "Cloud Infrastructure": 44.4,
     "Product & UX": 33.3,
     "Memory & Tools": 61.9,
     "Hard Fixes": 16.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 100.0,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 83.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 55.3
    }
   },
   "muse-spark-open": {
    "scores": {
     "LLM API Integration": 33.3,
     "Agent Loop": 66.7,
     "Billing & Limits": 33.3,
     "Cloud Infrastructure": 83.3,
     "Product & UX": 58.3,
     "Memory & Tools": 86.9,
     "Hard Fixes": 35.8
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "LLM API Integration": 50.0,
     "Agent Loop": 66.7,
     "Billing & Limits": 94.4,
     "Cloud Infrastructure": 100.0,
     "Product & UX": 91.7,
     "Memory & Tools": 66.7,
     "Hard Fixes": 59.7
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "LLM API Integration": 55.6,
     "Agent Loop": 0.0,
     "Billing & Limits": 33.3,
     "Cloud Infrastructure": 83.3,
     "Product & UX": 0.0,
     "Memory & Tools": 86.9,
     "Hard Fixes": 11.7
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "LLM API Integration": 55.6,
     "Agent Loop": 50.0,
     "Billing & Limits": 61.1,
     "Cloud Infrastructure": 61.1,
     "Product & UX": 50.0,
     "Memory & Tools": 95.2,
     "Hard Fixes": 26.0
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "LLM API Integration": 83.3,
     "Agent Loop": 100.0,
     "Billing & Limits": 72.2,
     "Cloud Infrastructure": 72.2,
     "Product & UX": 58.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 29.4
    }
   },
   "qwen3.8-max": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 100.0,
     "Billing & Limits": 61.1,
     "Cloud Infrastructure": 58.3,
     "Product & UX": 50.0,
     "Memory & Tools": 66.7,
     "Hard Fixes": 46.9
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "LLM API Integration": 66.7,
     "Agent Loop": 16.7,
     "Billing & Limits": 33.3,
     "Cloud Infrastructure": 61.1,
     "Product & UX": 75.0,
     "Memory & Tools": 86.9,
     "Hard Fixes": 28.3
    }
   },
   "step-5-preview": {
    "scores": {
     "LLM API Integration": 100.0,
     "Agent Loop": 100.0,
     "Billing & Limits": 61.1,
     "Cloud Infrastructure": 55.6,
     "Product & UX": 58.3,
     "Memory & Tools": 100.0,
     "Hard Fixes": 36.3
    }
   }
  }
 },
 "Browser": {
  "parts": [
   {
    "name": "Data Aggregation",
    "gloss": "Exhaustive traversal of a large record set and multi-part roll-ups.",
    "tasks": 6
   },
   {
    "name": "Sequential Planning",
    "gloss": "Order-dependent steps where each step constrains the next.",
    "tasks": 4
   },
   {
    "name": "Analysis-Driven Action",
    "gloss": "Derive a result from the data, then act on it in the app.",
    "tasks": 7
   },
   {
    "name": "Structural Integrity",
    "gloss": "Keep a document or record graph consistent across an edit.",
    "tasks": 5
   },
   {
    "name": "Constraint Adherence",
    "gloss": "Exact fidelity, plus correctly declining what the app won't permit.",
    "tasks": 3
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Data Aggregation": 83.3,
     "Sequential Planning": 71.4,
     "Analysis-Driven Action": 97.8,
     "Structural Integrity": 94.8,
     "Constraint Adherence": 100.0
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Data Aggregation": 72.8,
     "Sequential Planning": 71.4,
     "Analysis-Driven Action": 97.3,
     "Structural Integrity": 95.6,
     "Constraint Adherence": 97.2
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Data Aggregation": 40.5,
     "Sequential Planning": 57.4,
     "Analysis-Driven Action": 88.8,
     "Structural Integrity": 91.7,
     "Constraint Adherence": 61.8
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Data Aggregation": 99.2,
     "Sequential Planning": 69.6,
     "Analysis-Driven Action": 80.7,
     "Structural Integrity": 95.6,
     "Constraint Adherence": 97.9
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Data Aggregation": 82.6,
     "Sequential Planning": 69.6,
     "Analysis-Driven Action": 96.4,
     "Structural Integrity": 97.3,
     "Constraint Adherence": 99.3
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Data Aggregation": 66.7,
     "Sequential Planning": 89.3,
     "Analysis-Driven Action": 56.5,
     "Structural Integrity": 52.1,
     "Constraint Adherence": 88.9
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Data Aggregation": 39.4,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 47.5,
     "Structural Integrity": 53.8,
     "Constraint Adherence": 96.5
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Data Aggregation": 83.3,
     "Sequential Planning": 83.9,
     "Analysis-Driven Action": 74.5,
     "Structural Integrity": 86.1,
     "Constraint Adherence": 93.8
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Data Aggregation": 64.3,
     "Sequential Planning": 70.2,
     "Analysis-Driven Action": 77.3,
     "Structural Integrity": 98.6,
     "Constraint Adherence": 100.0
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Data Aggregation": 0.0,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 0.0,
     "Structural Integrity": 0.0,
     "Constraint Adherence": 0.0
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Data Aggregation": 48.9,
     "Sequential Planning": 54.6,
     "Analysis-Driven Action": 65.5,
     "Structural Integrity": 48.9,
     "Constraint Adherence": 93.8
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Data Aggregation": 56.1,
     "Sequential Planning": 78.7,
     "Analysis-Driven Action": 76.9,
     "Structural Integrity": 57.0,
     "Constraint Adherence": 94.4
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Data Aggregation": 83.3,
     "Sequential Planning": 96.4,
     "Analysis-Driven Action": 97.1,
     "Structural Integrity": 83.6,
     "Constraint Adherence": 100.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Data Aggregation": 72.8,
     "Sequential Planning": 83.2,
     "Analysis-Driven Action": 58.4,
     "Structural Integrity": 57.3,
     "Constraint Adherence": 97.9
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Data Aggregation": 83.3,
     "Sequential Planning": 96.4,
     "Analysis-Driven Action": 97.1,
     "Structural Integrity": 93.1,
     "Constraint Adherence": 100.0
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Data Aggregation": 91.3,
     "Sequential Planning": 93.8,
     "Analysis-Driven Action": 90.4,
     "Structural Integrity": 94.6,
     "Constraint Adherence": 100.0
    }
   },
   "grok-4.6": {
    "scores": {
     "Data Aggregation": 17.6,
     "Sequential Planning": 14.8,
     "Analysis-Driven Action": 59.3,
     "Structural Integrity": 72.9,
     "Constraint Adherence": 96.5
    }
   },
   "grok-4.7": {
    "scores": {
     "Data Aggregation": 68.2,
     "Sequential Planning": 17.5,
     "Analysis-Driven Action": 6.3,
     "Structural Integrity": 97.3,
     "Constraint Adherence": 74.7
    }
   },
   "inkling": {
    "scores": {
     "Data Aggregation": 50.0,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 0.0,
     "Structural Integrity": 0.0,
     "Constraint Adherence": 0.0
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Data Aggregation": 83.3,
     "Sequential Planning": 91.3,
     "Analysis-Driven Action": 63.2,
     "Structural Integrity": 69.0,
     "Constraint Adherence": 81.9
    }
   },
   "kimi-k3": {
    "scores": {
     "Data Aggregation": 66.7,
     "Sequential Planning": 59.2,
     "Analysis-Driven Action": 88.6,
     "Structural Integrity": 75.5,
     "Constraint Adherence": 81.6
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Data Aggregation": 57.4,
     "Sequential Planning": 67.1,
     "Analysis-Driven Action": 81.1,
     "Structural Integrity": 94.8,
     "Constraint Adherence": 93.1
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Data Aggregation": 82.6,
     "Sequential Planning": 69.4,
     "Analysis-Driven Action": 80.7,
     "Structural Integrity": 76.5,
     "Constraint Adherence": 98.6
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Data Aggregation": 4.9,
     "Sequential Planning": 8.8,
     "Analysis-Driven Action": 0.0,
     "Structural Integrity": 54.0,
     "Constraint Adherence": 33.3
    }
   },
   "mistral-large-4": {
    "scores": {
     "Data Aggregation": 0.0,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 0.0,
     "Structural Integrity": 3.0,
     "Constraint Adherence": 0.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Data Aggregation": 35.2,
     "Sequential Planning": 25.0,
     "Analysis-Driven Action": 6.2,
     "Structural Integrity": 33.6,
     "Constraint Adherence": 83.3
    }
   },
   "muse-spark-open": {
    "scores": {
     "Data Aggregation": 34.7,
     "Sequential Planning": 14.8,
     "Analysis-Driven Action": 17.6,
     "Structural Integrity": 21.5,
     "Constraint Adherence": 52.1
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Data Aggregation": 22.0,
     "Sequential Planning": 71.4,
     "Analysis-Driven Action": 58.9,
     "Structural Integrity": 75.5,
     "Constraint Adherence": 97.9
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Data Aggregation": 0.7,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 15.8,
     "Structural Integrity": 13.7,
     "Constraint Adherence": 89.1
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Data Aggregation": 17.8,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 24.3,
     "Structural Integrity": 21.7,
     "Constraint Adherence": 51.4
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Data Aggregation": 33.3,
     "Sequential Planning": 7.4,
     "Analysis-Driven Action": 15.7,
     "Structural Integrity": 36.7,
     "Constraint Adherence": 27.8
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Data Aggregation": 39.9,
     "Sequential Planning": 31.6,
     "Analysis-Driven Action": 59.0,
     "Structural Integrity": 45.3,
     "Constraint Adherence": 76.0
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Data Aggregation": 1.1,
     "Sequential Planning": 0.0,
     "Analysis-Driven Action": 35.8,
     "Structural Integrity": 32.1,
     "Constraint Adherence": 74.1
    }
   },
   "step-5-preview": {
    "scores": {
     "Data Aggregation": 18.0,
     "Sequential Planning": 3.4,
     "Analysis-Driven Action": 17.9,
     "Structural Integrity": 45.2,
     "Constraint Adherence": 49.3
    }
   }
  }
 },
 "Data Analysis": {
  "parts": [
   {
    "name": "Consecutive events",
    "gloss": "Find users or entities with runs of consecutive growth in a small, messy table of dated values.",
    "tasks": 47
   },
   {
    "name": "Table join",
    "gloss": "Map the columns of one table onto the obfuscated columns of another.",
    "tasks": 52
   },
   {
    "name": "Table reformat",
    "gloss": "Convert a table between CSV, TSV, JSON, JSONL, Markdown and HTML.",
    "tasks": 51
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Consecutive events": 90.6,
     "Table join": 56.2,
     "Table reformat": 94.1
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Consecutive events": 90.6,
     "Table join": 55.8,
     "Table reformat": 98.0
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Consecutive events": 16.4,
     "Table join": 44.8,
     "Table reformat": 98.0
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Consecutive events": 90.6,
     "Table join": 52.3,
     "Table reformat": 98.0
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Consecutive events": 40.6,
     "Table join": 39.8,
     "Table reformat": 98.0
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Consecutive events": 90.3,
     "Table join": 48.2,
     "Table reformat": 100.0
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Consecutive events": 20.6,
     "Table join": 45.3,
     "Table reformat": 96.1
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Consecutive events": 91.4,
     "Table join": 69.9,
     "Table reformat": 94.1
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Consecutive events": 98.2,
     "Table join": 67.3,
     "Table reformat": 94.1
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Consecutive events": 13.6,
     "Table join": 46.3,
     "Table reformat": 100.0
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Consecutive events": 84.3,
     "Table join": 48.8,
     "Table reformat": 96.1
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Consecutive events": 87.7,
     "Table join": 50.2,
     "Table reformat": 100.0
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Consecutive events": 90.6,
     "Table join": 58.4,
     "Table reformat": 100.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Consecutive events": 70.5,
     "Table join": 49.6,
     "Table reformat": 100.0
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Consecutive events": 90.6,
     "Table join": 57.4,
     "Table reformat": 100.0
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Consecutive events": 90.5,
     "Table join": 56.5,
     "Table reformat": 100.0
    }
   },
   "grok-4.6": {
    "scores": {
     "Consecutive events": 84.6,
     "Table join": 43.1,
     "Table reformat": 98.0
    }
   },
   "grok-4.7": {
    "scores": {
     "Consecutive events": 85.7,
     "Table join": 44.9,
     "Table reformat": 100.0
    }
   },
   "inkling": {
    "scores": {
     "Consecutive events": 69.9,
     "Table join": 48.5,
     "Table reformat": 100.0
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Consecutive events": 90.2,
     "Table join": 49.1,
     "Table reformat": 98.0
    }
   },
   "kimi-k3": {
    "scores": {
     "Consecutive events": 89.8,
     "Table join": 48.4,
     "Table reformat": 98.0
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Consecutive events": 70.5,
     "Table join": 48.3,
     "Table reformat": 98.0
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Consecutive events": 68.1,
     "Table join": 48.0,
     "Table reformat": 98.0
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Consecutive events": 81.6,
     "Table join": 48.2,
     "Table reformat": 100.0
    }
   },
   "mistral-large-4": {
    "scores": {
     "Consecutive events": 80.5,
     "Table join": 49.1,
     "Table reformat": 100.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Consecutive events": 90.5,
     "Table join": 48.2,
     "Table reformat": 100.0
    }
   },
   "muse-spark-open": {
    "scores": {
     "Consecutive events": 36.0,
     "Table join": 44.5,
     "Table reformat": 100.0
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Consecutive events": 98.5,
     "Table join": 46.3,
     "Table reformat": 100.0
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Consecutive events": 50.4,
     "Table join": 46.7,
     "Table reformat": 100.0
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Consecutive events": 49.2,
     "Table join": 46.4,
     "Table reformat": 98.0
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Consecutive events": 84.9,
     "Table join": 48.8,
     "Table reformat": 96.1
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Consecutive events": 87.1,
     "Table join": 50.1,
     "Table reformat": 98.0
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Consecutive events": 50.3,
     "Table join": 45.1,
     "Table reformat": 100.0
    }
   },
   "step-5-preview": {
    "scores": {
     "Consecutive events": 42.4,
     "Table join": 50.1,
     "Table reformat": 98.0
    }
   }
  }
 },
 "Research": {
  "parts": [
   {
    "name": "Regulatory & Compliance",
    "gloss": "Rules, filings and compliance questions answered from primary sources.",
    "tasks": 7
   },
   {
    "name": "Legal",
    "gloss": "Statutes, cases and contractual questions with cited authority.",
    "tasks": 4
   },
   {
    "name": "Tax & Finance",
    "gloss": "Tax rules and financial figures, with the arithmetic shown.",
    "tasks": 5
   },
   {
    "name": "Strategy",
    "gloss": "Market and competitive analysis with sourced evidence.",
    "tasks": 3
   },
   {
    "name": "Science & Evidence",
    "gloss": "Scientific and empirical questions weighed on the published evidence.",
    "tasks": 3
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Regulatory & Compliance": 94.2,
     "Legal": 99.2,
     "Tax & Finance": 82.2,
     "Strategy": 66.5,
     "Science & Evidence": 93.5
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Regulatory & Compliance": 95.3,
     "Legal": 99.2,
     "Tax & Finance": 94.1,
     "Strategy": 93.5,
     "Science & Evidence": 88.2
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Regulatory & Compliance": 80.1,
     "Legal": 90.1,
     "Tax & Finance": 76.1,
     "Strategy": 91.9,
     "Science & Evidence": 92.2
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Regulatory & Compliance": 92.2,
     "Legal": 99.9,
     "Tax & Finance": 91.1,
     "Strategy": 89.3,
     "Science & Evidence": 62.4
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Regulatory & Compliance": 96.7,
     "Legal": 96.2,
     "Tax & Finance": 64.4,
     "Strategy": 89.1,
     "Science & Evidence": 94.8
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Regulatory & Compliance": 80.3,
     "Legal": 67.0,
     "Tax & Finance": 78.2,
     "Strategy": 80.4,
     "Science & Evidence": 89.0
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Regulatory & Compliance": 78.0,
     "Legal": 91.5,
     "Tax & Finance": 74.2,
     "Strategy": 77.1,
     "Science & Evidence": 77.6
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Regulatory & Compliance": 84.7,
     "Legal": 99.3,
     "Tax & Finance": 81.8,
     "Strategy": 83.5,
     "Science & Evidence": 86.7
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Regulatory & Compliance": 94.9,
     "Legal": 98.8,
     "Tax & Finance": 91.7,
     "Strategy": 89.2,
     "Science & Evidence": 83.4
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Regulatory & Compliance": 29.5,
     "Legal": 34.3,
     "Tax & Finance": 54.6,
     "Strategy": 46.5,
     "Science & Evidence": 27.3
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Regulatory & Compliance": 80.9,
     "Legal": 72.1,
     "Tax & Finance": 55.2,
     "Strategy": 75.6,
     "Science & Evidence": 89.4
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Regulatory & Compliance": 85.9,
     "Legal": 99.7,
     "Tax & Finance": 58.3,
     "Strategy": 65.6,
     "Science & Evidence": 88.6
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Regulatory & Compliance": 87.9,
     "Legal": 94.3,
     "Tax & Finance": 93.4,
     "Strategy": 93.6,
     "Science & Evidence": 84.0
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Regulatory & Compliance": 51.0,
     "Legal": 72.6,
     "Tax & Finance": 13.6,
     "Strategy": 58.4,
     "Science & Evidence": 58.3
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Regulatory & Compliance": 84.1,
     "Legal": 99.7,
     "Tax & Finance": 92.1,
     "Strategy": 90.5,
     "Science & Evidence": 86.7
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Regulatory & Compliance": 73.7,
     "Legal": 91.9,
     "Tax & Finance": 69.6,
     "Strategy": 79.7,
     "Science & Evidence": 89.8
    }
   },
   "grok-4.6": {
    "scores": {
     "Regulatory & Compliance": 90.5,
     "Legal": 93.2,
     "Tax & Finance": 85.8,
     "Strategy": 85.9,
     "Science & Evidence": 86.8
    }
   },
   "grok-4.7": {
    "scores": {
     "Regulatory & Compliance": 89.1,
     "Legal": 92.9,
     "Tax & Finance": 86.5,
     "Strategy": 91.6,
     "Science & Evidence": 88.9
    }
   },
   "inkling": {
    "scores": {
     "Regulatory & Compliance": 59.3,
     "Legal": 56.0,
     "Tax & Finance": 64.2,
     "Strategy": 63.6,
     "Science & Evidence": 58.9
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Regulatory & Compliance": 89.5,
     "Legal": 98.4,
     "Tax & Finance": 66.8,
     "Strategy": 92.4,
     "Science & Evidence": 58.4
    }
   },
   "kimi-k3": {
    "scores": {
     "Regulatory & Compliance": 87.4,
     "Legal": 93.7,
     "Tax & Finance": 64.6,
     "Strategy": 61.1,
     "Science & Evidence": 87.5
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Regulatory & Compliance": 84.8,
     "Legal": 88.3,
     "Tax & Finance": 72.7,
     "Strategy": 93.6,
     "Science & Evidence": 91.6
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Regulatory & Compliance": 68.9,
     "Legal": 78.7,
     "Tax & Finance": 68.4,
     "Strategy": 88.8,
     "Science & Evidence": 94.3
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Regulatory & Compliance": 75.2,
     "Legal": 84.5,
     "Tax & Finance": 76.8,
     "Strategy": 84.4,
     "Science & Evidence": 88.4
    }
   },
   "mistral-large-4": {
    "scores": {
     "Regulatory & Compliance": 38.0,
     "Legal": 36.0,
     "Tax & Finance": 9.9,
     "Strategy": 50.5,
     "Science & Evidence": 22.0
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Regulatory & Compliance": 94.8,
     "Legal": 97.6,
     "Tax & Finance": 60.9,
     "Strategy": 70.1,
     "Science & Evidence": 81.8
    }
   },
   "muse-spark-open": {
    "scores": {
     "Regulatory & Compliance": 79.5,
     "Legal": 89.4,
     "Tax & Finance": 80.0,
     "Strategy": 82.6,
     "Science & Evidence": 85.7
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Regulatory & Compliance": 90.7,
     "Legal": 99.1,
     "Tax & Finance": 88.6,
     "Strategy": 98.9,
     "Science & Evidence": 92.1
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Regulatory & Compliance": 54.8,
     "Legal": 78.9,
     "Tax & Finance": 67.4,
     "Strategy": 67.2,
     "Science & Evidence": 81.7
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Regulatory & Compliance": 69.5,
     "Legal": 71.6,
     "Tax & Finance": 61.5,
     "Strategy": 75.0,
     "Science & Evidence": 56.7
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Regulatory & Compliance": 65.2,
     "Legal": 74.2,
     "Tax & Finance": 36.9,
     "Strategy": 74.0,
     "Science & Evidence": 62.8
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Regulatory & Compliance": 82.1,
     "Legal": 87.3,
     "Tax & Finance": 57.2,
     "Strategy": 76.0,
     "Science & Evidence": 55.0
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Regulatory & Compliance": 50.6,
     "Legal": 71.8,
     "Tax & Finance": 68.7,
     "Strategy": 81.7,
     "Science & Evidence": 79.9
    }
   },
   "step-5-preview": {
    "scores": {
     "Regulatory & Compliance": 71.7,
     "Legal": 71.3,
     "Tax & Finance": 74.4,
     "Strategy": 83.3,
     "Science & Evidence": 84.3
    }
   }
  }
 },
 "Design": {
  "parts": [
   {
    "name": "Documents (DOCX)",
    "gloss": "Word documents: layout, structure and formatting to a brief.",
    "tasks": 5
   },
   {
    "name": "PDF",
    "gloss": "PDF deliverables: typeset pages, forms and reports.",
    "tasks": 8
   },
   {
    "name": "Presentations (PPT)",
    "gloss": "Slide decks: visual design, content organization and fit to the request.",
    "tasks": 12
   }
  ],
  "models": {
   "claude-fable-5.1": {
    "scores": {
     "Documents (DOCX)": 89.8,
     "PDF": 84.1,
     "Presentations (PPT)": 76.4
    }
   },
   "claude-fable-5.2": {
    "scores": {
     "Documents (DOCX)": 92.6,
     "PDF": 87.2,
     "Presentations (PPT)": 80.5
    }
   },
   "claude-haiku-5.5": {
    "scores": {
     "Documents (DOCX)": 90.1,
     "PDF": 84.6,
     "Presentations (PPT)": 71.2
    }
   },
   "claude-opus-5.5": {
    "scores": {
     "Documents (DOCX)": 86.0,
     "PDF": 85.6,
     "Presentations (PPT)": 80.7
    }
   },
   "claude-sonnet-5.5": {
    "scores": {
     "Documents (DOCX)": 92.1,
     "PDF": 86.4,
     "Presentations (PPT)": 73.4
    }
   },
   "deepseek-v4.1-flash-1017": {
    "scores": {
     "Documents (DOCX)": 79.8,
     "PDF": 86.7,
     "Presentations (PPT)": 75.2
    }
   },
   "gemini-3.8-flash": {
    "scores": {
     "Documents (DOCX)": 80.3,
     "PDF": 72.3,
     "Presentations (PPT)": 63.5
    }
   },
   "gemini-4-argon": {
    "scores": {
     "Documents (DOCX)": 82.2,
     "PDF": 69.8,
     "Presentations (PPT)": 63.4
    }
   },
   "gemini-4-carbon": {
    "scores": {
     "Documents (DOCX)": 92.1,
     "PDF": 86.1,
     "Presentations (PPT)": 71.2
    }
   },
   "gemma-4-31b": {
    "scores": {
     "Documents (DOCX)": 61.1,
     "PDF": 52.7,
     "Presentations (PPT)": 57.2
    }
   },
   "glm-5.3-flash": {
    "scores": {
     "Documents (DOCX)": 70.8,
     "PDF": 77.1,
     "Presentations (PPT)": 72.6
    }
   },
   "gpt-5.6-terra": {
    "scores": {
     "Documents (DOCX)": 76.4,
     "PDF": 70.8,
     "Presentations (PPT)": 68.6
    }
   },
   "gpt-6-astra": {
    "scores": {
     "Documents (DOCX)": 81.6,
     "PDF": 78.3,
     "Presentations (PPT)": 80.5
    }
   },
   "gpt-6-luna": {
    "scores": {
     "Documents (DOCX)": 81.9,
     "PDF": 77.4,
     "Presentations (PPT)": 72.8
    }
   },
   "gpt-6.1-sol": {
    "scores": {
     "Documents (DOCX)": 83.3,
     "PDF": 78.9,
     "Presentations (PPT)": 78.4
    }
   },
   "gpt-6.1-terra": {
    "scores": {
     "Documents (DOCX)": 89.4,
     "PDF": 74.9,
     "Presentations (PPT)": 78.8
    }
   },
   "grok-4.6": {
    "scores": {
     "Documents (DOCX)": 73.5,
     "PDF": 80.8,
     "Presentations (PPT)": 74.4
    }
   },
   "grok-4.7": {
    "scores": {
     "Documents (DOCX)": 72.0,
     "PDF": 79.3,
     "Presentations (PPT)": 73.3
    }
   },
   "inkling": {
    "scores": {
     "Documents (DOCX)": 65.1,
     "PDF": 47.7,
     "Presentations (PPT)": 56.1
    }
   },
   "kimi-k3.1-pro": {
    "scores": {
     "Documents (DOCX)": 76.8,
     "PDF": 80.3,
     "Presentations (PPT)": 68.0
    }
   },
   "kimi-k3": {
    "scores": {
     "Documents (DOCX)": 77.3,
     "PDF": 76.8,
     "Presentations (PPT)": 67.7
    }
   },
   "mimo-v2.6-pro": {
    "scores": {
     "Documents (DOCX)": 82.1,
     "PDF": 76.9,
     "Presentations (PPT)": 69.2
    }
   },
   "minimax-m3.1-flash": {
    "scores": {
     "Documents (DOCX)": 82.3,
     "PDF": 76.1,
     "Presentations (PPT)": 63.4
    }
   },
   "mistral-large-4-0": {
    "scores": {
     "Documents (DOCX)": 67.3,
     "PDF": 79.1,
     "Presentations (PPT)": 78.2
    }
   },
   "mistral-large-4": {
    "scores": {
     "Documents (DOCX)": 77.7,
     "PDF": 50.1,
     "Presentations (PPT)": 60.6
    }
   },
   "muse-spark-1.3": {
    "scores": {
     "Documents (DOCX)": 80.8,
     "PDF": 80.8,
     "Presentations (PPT)": 68.4
    }
   },
   "muse-spark-open": {
    "scores": {
     "Documents (DOCX)": 60.4,
     "PDF": 64.4,
     "Presentations (PPT)": 52.9
    }
   },
   "muse-wonder-1.0": {
    "scores": {
     "Documents (DOCX)": 70.6,
     "PDF": 70.2,
     "Presentations (PPT)": 71.5
    }
   },
   "nemotron-super-vl-3.5-nvfp4": {
    "scores": {
     "Documents (DOCX)": 59.4,
     "PDF": 57.9,
     "Presentations (PPT)": 52.6
    }
   },
   "nemotron-super-vl-3.5": {
    "scores": {
     "Documents (DOCX)": 70.1,
     "PDF": 58.9,
     "Presentations (PPT)": 56.4
    }
   },
   "qwen3.8-27b": {
    "scores": {
     "Documents (DOCX)": 76.9,
     "PDF": 78.8,
     "Presentations (PPT)": 66.3
    }
   },
   "qwen3.8-max": {
    "scores": {
     "Documents (DOCX)": 77.3,
     "PDF": 80.3,
     "Presentations (PPT)": 73.9
    }
   },
   "smaug-nemotron-3.5-nvfp4": {
    "scores": {
     "Documents (DOCX)": 66.3,
     "PDF": 59.9,
     "Presentations (PPT)": 58.7
    }
   },
   "step-5-preview": {
    "scores": {
     "Documents (DOCX)": 78.2,
     "PDF": 69.2,
     "Presentations (PPT)": 68.9
    }
   }
  }
 }
}