{
  "checked_on": "2026-10-26",
  "currency": "USD",
  "mode": "standard API-equivalent token estimates",
  "formula": "Sum each request's uncached input, cached input and total output at its applicable per-million rates. Reasoning is a subset of output. Cache writes are zero in these records.",
  "openai_long_context_rule": "For the Sol and Luna model designations costed here: above 272000 input tokens, apply 2x input/cache and 1.5x output to the full request.",
  "sources": [
    "https://developers.openai.com/api/docs/pricing",
    "https://developers.openai.com/api/docs/models/gpt-6.1-sol",
    "https://developers.openai.com/api/docs/models/gpt-6-luna",
    "https://api-docs.deepseek.com/quick_start/pricing/"
  ],
  "exclusions": [
    "setup/preflight",
    "tools",
    "coordination and publication",
    "unpersisted verifier/background usage",
    "taxes",
    "subscription allocations",
    "invoice reconciliation"
  ],
  "notes": [
    "Recorded cache counters, not estimated cache-hit rates.",
    "Different providers/tokenizers and harnesses limit comparison.",
    "No acceptance-based cost per published article or customer savings was measured."
  ],
  "runs": [
    {
      "scope": "sol",
      "model": "gpt-6.1-sol",
      "request_count": 30,
      "long_context_requests": 0,
      "tokens": {
        "input_tokens": 3046820,
        "cached_input_tokens": 2894720,
        "cache_write_input_tokens": 0,
        "output_tokens": 41141,
        "reasoning_output_tokens": 5500,
        "total_tokens": 3087961
      },
      "rates_usd_per_million_input_cache_output": [
        "2",
        "0.1",
        "10"
      ],
      "estimated_usd": "1.0050820"
    },
    {
      "scope": "luna",
      "model": "gpt-6-luna",
      "request_count": 43,
      "long_context_requests": 0,
      "tokens": {
        "input_tokens": 3110420,
        "cached_input_tokens": 3009152,
        "cache_write_input_tokens": 0,
        "output_tokens": 47354,
        "reasoning_output_tokens": 9353,
        "total_tokens": 3157774
      },
      "rates_usd_per_million_input_cache_output": [
        "0.1",
        "0.01",
        "0.5"
      ],
      "estimated_usd": "0.06389532"
    },
    {
      "scope": "three-way-review",
      "model": "gpt-6.1-sol",
      "request_count": 65,
      "long_context_requests": 26,
      "tokens": {
        "input_tokens": 14680963,
        "cached_input_tokens": 14342272,
        "cache_write_input_tokens": 0,
        "output_tokens": 54873,
        "reasoning_output_tokens": 15308,
        "total_tokens": 14735836
      },
      "rates_usd_per_million_input_cache_output": [
        "2",
        "0.1",
        "10"
      ],
      "estimated_usd": "3.9670558"
    },
    {
      "scope": "deepseek",
      "model": "deepseek/deepseek-v4.1-flash (latest; served snapshot unverified)",
      "request_count": 105,
      "tokens": {
        "input_tokens": 31878376,
        "cached_input_tokens": 31187584,
        "cache_write_input_tokens": 0,
        "output_tokens": 177989,
        "reasoning_output_tokens": null,
        "total_tokens": 32056365,
        "uncached_input_tokens": 690792
      },
      "rates_usd_per_million_input_cache_output": [
        "0.15",
        "0.003",
        "0.6"
      ],
      "estimated_usd": "0.303974952",
      "pricing_basis": "DeepSeek direct API off-peak reference rates; Sunday 25 October UTC; not a CommandCode invoice"
    }
  ]
}