{
  "scope": "Solver inference only, API-equivalent at list rates; the model ran on a ChatGPT subscription through Codex, so these are estimates, not bills. Judging and local infrastructure are excluded. Every run in the sweep is priced.",
  "pricing_verified": "2026-10-26",
  "sources": {
    "openai": "https://developers.openai.com/api/docs/pricing",
    "model_page": "https://developers.openai.com/api/docs/models/gpt-6.1-terra"
  },
  "rates_per_million": {
    "gpt61terra": {
      "input": 1.0,
      "cached_input": 0.05,
      "output": 5.0
    }
  },
  "groups": [
    {
      "model": "gpt61terra",
      "effort": "low",
      "n": 23,
      "usd": {
        "n": 23,
        "mean": 0.11388121739130434,
        "se": 0.012431742121271202
      },
      "usd_total": 2.6192680000000004,
      "raw_tokens": {
        "n": 23,
        "mean": 772326.9130434783,
        "se": 124553.3036918452
      },
      "raw_tokens_total": 17763519,
      "output_tokens": {
        "n": 23,
        "mean": 5475.260869565217,
        "se": 514.1837470792324
      },
      "reasoning_output_tokens": {
        "n": 23,
        "mean": 957.2608695652174,
        "se": 212.33443880523112
      },
      "cached_input_share": {
        "n": 23,
        "mean": 0.9262798511394646,
        "se": 0.004215773617243187
      },
      "minutes": {
        "n": 23,
        "mean": 2.784699275362319,
        "se": 0.3558717242999399
      },
      "solver_fallback_runs": 0
    },
    {
      "model": "gpt61terra",
      "effort": "medium",
      "n": 23,
      "usd": {
        "n": 23,
        "mean": 0.13880178260869566,
        "se": 0.0157420831414473
      },
      "usd_total": 3.1924410000000005,
      "raw_tokens": {
        "n": 23,
        "mean": 980738.1739130435,
        "se": 151330.81756814927
      },
      "raw_tokens_total": 22556978,
      "output_tokens": {
        "n": 23,
        "mean": 7896.260869565217,
        "se": 849.023184870818
      },
      "reasoning_output_tokens": {
        "n": 23,
        "mean": 1892.4782608695652,
        "se": 368.2986783583332
      },
      "cached_input_share": {
        "n": 23,
        "mean": 0.9353193496044325,
        "se": 0.004855416309227155
      },
      "minutes": {
        "n": 23,
        "mean": 3.112063768115942,
        "se": 0.3568523321664143
      },
      "solver_fallback_runs": 0
    },
    {
      "model": "gpt61terra",
      "effort": "high",
      "n": 23,
      "usd": {
        "n": 23,
        "mean": 0.12833173913043477,
        "se": 0.009959278123636384
      },
      "usd_total": 2.9516300000000006,
      "raw_tokens": {
        "n": 23,
        "mean": 901051.9565217391,
        "se": 109686.98253594103
      },
      "raw_tokens_total": 20724195,
      "output_tokens": {
        "n": 23,
        "mean": 7152,
        "se": 435.6344471474932
      },
      "reasoning_output_tokens": {
        "n": 23,
        "mean": 1417.9130434782608,
        "se": 135.15920802522427
      },
      "cached_input_share": {
        "n": 23,
        "mean": 0.9381341315379912,
        "se": 0.003469340527588736
      },
      "minutes": {
        "n": 23,
        "mean": 3.204827536231884,
        "se": 0.3749792218465635
      },
      "solver_fallback_runs": 0
    },
    {
      "model": "gpt61terra",
      "effort": "extra-high",
      "n": 23,
      "usd": {
        "n": 23,
        "mean": 0.24049547826086956,
        "se": 0.05293864630337905
      },
      "usd_total": 5.531395999999999,
      "raw_tokens": {
        "n": 23,
        "mean": 2154061.565217391,
        "se": 663391.7521656494
      },
      "raw_tokens_total": 49543416,
      "output_tokens": {
        "n": 23,
        "mean": 12930.652173913044,
        "se": 2541.31760867796
      },
      "reasoning_output_tokens": {
        "n": 23,
        "mean": 3735.1304347826085,
        "se": 1250.7415544330352
      },
      "cached_input_share": {
        "n": 23,
        "mean": 0.9493486907166425,
        "se": 0.003906255905420619
      },
      "minutes": {
        "n": 23,
        "mean": 4.949247826086957,
        "se": 0.8244009545206528
      },
      "solver_fallback_runs": 0
    },
    {
      "model": "gpt61terra",
      "effort": "max",
      "n": 23,
      "usd": {
        "n": 23,
        "mean": 0.356363347826087,
        "se": 0.03309458686697399
      },
      "usd_total": 8.196356999999999,
      "raw_tokens": {
        "n": 23,
        "mean": 2377189.5217391304,
        "se": 388193.14662501693
      },
      "raw_tokens_total": 54675359,
      "output_tokens": {
        "n": 23,
        "mean": 28762.82608695652,
        "se": 1362.0329255656043
      },
      "reasoning_output_tokens": {
        "n": 23,
        "mean": 15331.739130434782,
        "se": 742.0232708319141
      },
      "cached_input_share": {
        "n": 23,
        "mean": 0.9481161433001566,
        "se": 0.004519520865675356
      },
      "minutes": {
        "n": 23,
        "mean": 6.877489855072463,
        "se": 0.5087577231911165
      },
      "solver_fallback_runs": 0
    }
  ],
  "totals": {
    "gpt61terra": {
      "runs": 115,
      "usd": 22.491092000000005,
      "raw_tokens": 165263467,
      "solver_hours": 8.022525833333335
    }
  },
  "limitations": [
    "Codex receipts report input, cached input, output and reasoning tokens per run; cached input is billed at the cache-read rate and the rest of the input at the standard rate. Reasoning tokens are part of output and are not billed twice.",
    "Standard tier, short-context rates; Codex keeps each request within its context window, so no long-context premium is modelled.",
    "No Batch, Flex, Fast or priority pricing is applied.",
    "The estimate covers the solver's exposed receipts, not an independently observed API invoice.",
    "The sweep stamped each run at run time at the published GPT-6.1 Terra list rates; the export recomputes every run from its receipt and matched every stamp."
  ],
  "comparison_sha256": "abf22133b3b2d89a1263c5ba0b64d22e37838e7554465c8513d5e87e8a431911",
  "note": "Per-run estimates are in runs.json (estimated_usd, raw_tokens, token_usage). Judging is excluded."
}