{
  "benchmark_id": "ai-editorial-20261022",
  "execution_mode": "goal",
  "reasoning_effort": "high",
  "notes": [
    "Latest runtime cumulative snapshot per pinned session; snapshots are never summed.",
    "Cached input is included in input; reasoning output is included in output where the provider reports it that way. These subset fields must not be added again.",
    "Null means unavailable. No token figures are inferred from article word counts or subscription credits.",
    "Database tokens_used is retained only as a diagnostic; it is not substituted for cumulative usage.",
    "Elapsed time includes tools and waiting. Tool fees, unavailable internal work and monetary costs are not inferred.",
    "Completion marker records contestant readiness, not an independent quality judgment."
  ],
  "runs": [
    {
      "run_id": "sol",
      "configured_model": "gpt-6.1-sol",
      "elapsed_seconds": 1097.906772,
      "production_tokens": {
        "input_tokens": 3046820,
        "cached_input_tokens": 2894720,
        "cache_write_input_tokens": 0,
        "output_tokens": 41141,
        "reasoning_output_tokens": 5500,
        "total_tokens": 3087961
      },
      "usage_status": "runtime_reported"
    },
    {
      "run_id": "luna",
      "configured_model": "gpt-6-luna",
      "elapsed_seconds": 567.973218,
      "production_tokens": {
        "input_tokens": 3110420,
        "cached_input_tokens": 3009152,
        "cache_write_input_tokens": 0,
        "output_tokens": 47354,
        "reasoning_output_tokens": 9353,
        "total_tokens": 3157774
      },
      "usage_status": "runtime_reported"
    }
  ],
  "monetary_cost": null,
  "reviewer_tokens_separate_overhead": 8402872
}