{
  "schemaVersion": 1,
  "reviewedOn": "2026-10-30",
  "reviewedLabel": "October 30, 2026",
  "summary": "Start with a model suited to the step. Move up when better reasoning reduces mistakes or review effort.",
  "basis": "Suggested fit based on official provider documentation, not our own speed or quality benchmarks. Test on representative examples before relying on a model.",
  "pricingBasis": "USD per one million input and output tokens, using standard first-party API pricing. These are not subscription prices or total workflow costs. Tool calls, reasoning, retries, caching and different tokenizers affect the final bill.",
  "runtimeNote": "A model does not connect your tools or grant access. MCP support, available tools and model selection depend on the agent host and your organization's approved setup.",
  "models": {
    "gpt-6-luna": {
      "name": "GPT-6 Luna",
      "provider": "OpenAI",
      "apiId": "gpt-6-luna",
      "context": "1.05 million tokens",
      "providerPositioning": "Focused, cost-sensitive, high-volume work.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 0.1,
          "outputUsdPerMillion": 0.5,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://developers.openai.com/api/docs/models/gpt-6-luna",
      "pricingSourceUrl": "https://developers.openai.com/api/docs/pricing"
    },
    "gpt-6.1-sol": {
      "name": "GPT-6.1 Sol",
      "provider": "OpenAI",
      "apiId": "gpt-6.1-sol",
      "context": "1.05 million tokens",
      "providerPositioning": "Complex coding and professional work with a balance of intelligence and cost.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 2,
          "outputUsdPerMillion": 10,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://developers.openai.com/api/docs/models/gpt-6.1-sol",
      "pricingSourceUrl": "https://developers.openai.com/api/docs/pricing"
    },
    "gpt-6-astra": {
      "name": "GPT-6 Astra",
      "provider": "OpenAI",
      "apiId": "gpt-6-astra",
      "context": "1.05 million tokens",
      "providerPositioning": "Demanding reasoning and coding work.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 10,
          "outputUsdPerMillion": 50,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://developers.openai.com/api/docs/models/gpt-6-astra",
      "pricingSourceUrl": "https://developers.openai.com/api/docs/pricing"
    },
    "claude-haiku-5-5": {
      "name": "Claude Haiku 5.5",
      "provider": "Anthropic",
      "apiId": "claude-haiku-5-5",
      "context": "1 million tokens",
      "providerPositioning": "High-volume, latency-sensitive classification, extraction and routing.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 0.1,
          "outputUsdPerMillion": 0.5,
          "condition": "For requests with prompts up to 100,000 tokens."
        },
        {
          "label": "Standard API",
          "inputUsdPerMillion": 0.5,
          "outputUsdPerMillion": 2.5,
          "condition": "For requests with prompts over 100,000 tokens; the higher rates apply to that request."
        }
      ],
      "modelSourceUrl": "https://platform.claude.com/docs/en/models/haiku-5-5/overview",
      "pricingSourceUrl": "https://platform.claude.com/docs/en/about-claude/pricing"
    },
    "claude-sonnet-5-5": {
      "name": "Claude Sonnet 5.5",
      "provider": "Anthropic",
      "apiId": "claude-sonnet-5-5",
      "context": "1 million tokens",
      "providerPositioning": "A balance of speed and intelligence.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 2,
          "outputUsdPerMillion": 10,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://platform.claude.com/docs/en/models/sonnet-5-5/overview",
      "pricingSourceUrl": "https://platform.claude.com/docs/en/about-claude/pricing"
    },
    "claude-opus-5-5": {
      "name": "Claude Opus 5.5",
      "provider": "Anthropic",
      "apiId": "claude-opus-5-5",
      "context": "1 million tokens",
      "providerPositioning": "Long-running agentic coding and knowledge work.",
      "prices": [
        {
          "label": "Standard API",
          "inputUsdPerMillion": 4,
          "outputUsdPerMillion": 20,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://platform.claude.com/docs/en/models/opus-5-5/overview",
      "pricingSourceUrl": "https://platform.claude.com/docs/en/about-claude/pricing"
    },
    "gemini-3.5-flash-lite": {
      "name": "Gemini 3.5 Flash-Lite",
      "provider": "Google",
      "apiId": "gemini-3.5-flash-lite",
      "context": "1,048,576 input tokens",
      "providerPositioning": "Low-cost, high-throughput document parsing and simple extraction.",
      "prices": [
        {
          "label": "Standard paid API",
          "inputUsdPerMillion": 0.3,
          "outputUsdPerMillion": 2.5,
          "condition": ""
        }
      ],
      "modelSourceUrl": "https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash-lite",
      "pricingSourceUrl": "https://ai.google.dev/gemini-api/docs/pricing"
    }
  },
  "workflows": {
    "bounded-agent-work": {
      "start": {
        "modelId": "gpt-6.1-sol",
        "effort": "medium",
        "why": "A balanced starting point for a scoped code change and its acceptance checks."
      },
      "escalate": {
        "modelId": "gpt-6-astra",
        "when": "Difficult design tradeoffs or repeated meaningful failures justify more reasoning and a higher cost."
      },
      "alternatives": [
        {
          "modelId": "claude-sonnet-5-5",
          "when": "An alternative for the same bounded task in an approved Claude-based agent host; compare the resulting change and review effort."
        }
      ],
      "humanBoundary": "Keep scope, acceptance criteria and the decision to merge or deploy with the responsible person. More capable does not mean more authorized."
    },
    "knowledge-foundation": {
      "start": {
        "modelId": "claude-haiku-5-5",
        "effort": "",
        "why": "An economical starting point for extracting facts and references from a small, well-scoped set of sources."
      },
      "escalate": {
        "modelId": "claude-sonnet-5-5",
        "when": "Sources conflict, relationships span repositories, or the first draft needs substantial reasoning to reconcile the evidence."
      },
      "alternatives": [
        {
          "modelId": "gpt-6-luna",
          "when": "An economical extraction option in an approved OpenAI-based agent host."
        },
        {
          "modelId": "gemini-3.5-flash-lite",
          "when": "A document-parsing option when your approved setup already uses Gemini. Compare source accuracy on your own material."
        }
      ],
      "humanBoundary": "A larger context window does not establish completeness. Verify references, conflicts and unanswered questions before reusing the draft."
    },
    "grounded-support": {
      "start": {
        "modelId": "gpt-6-luna",
        "effort": "low",
        "why": "A low-cost starting point for a concise answer from a few relevant, approved source passages."
      },
      "escalate": {
        "modelId": "gpt-6.1-sol",
        "when": "The sources are present but the question requires complex diagnosis or reasoning across several system constraints."
      },
      "alternatives": [
        {
          "modelId": "claude-haiku-5-5",
          "when": "A volume-oriented option in an approved Claude-based support workflow."
        }
      ],
      "humanBoundary": "Missing, stale or conflicting evidence needs clarification or human help. A larger model must not invent a fix."
    },
    "support-improvement": {
      "start": {
        "modelId": "claude-sonnet-5-5",
        "effort": "",
        "why": "A balanced starting point for a minimal knowledge patch and analysis of replay results."
      },
      "escalate": {
        "modelId": "claude-opus-5-5",
        "when": "Contradictions, difficult regressions or changes to the knowledge structure require deeper investigation."
      },
      "alternatives": [
        {
          "modelId": "gpt-6.1-sol",
          "when": "An alternative for drafting and reviewing the proposal in an approved OpenAI-based agent host."
        },
        {
          "modelId": "claude-haiku-5-5",
          "when": "Use for initial thread classification or gap triage, then review the proposed patch with the stronger starting model."
        }
      ],
      "humanBoundary": "Run the replay comparison with actual tools or label it not run. Model judgment is not a test result; a named person still approves, revises or rejects publication."
    },
    "choose-ai-assistance": {
      "start": {
        "modelId": "gpt-6-luna",
        "effort": "low",
        "why": "An economical first pass through a structured worksheet; keep the experiment small and the output short."
      },
      "escalate": {
        "modelId": "gpt-6.1-sol",
        "when": "Ownership, growth opportunities and competing constraints need a more careful examination."
      },
      "alternatives": [
        {
          "modelId": "claude-haiku-5-5",
          "when": "An alternative for the first-pass plan in your approved Claude-based setup."
        }
      ],
      "humanBoundary": "The leader makes the choice. The model helps examine the tradeoffs; it does not decide whose development or responsibility to remove."
    },
    "daily-debrief": {
      "reviewedOn": "2026-11-02",
      "reviewedLabel": "November 2, 2026",
      "start": {
        "modelId": "gpt-6-luna",
        "effort": "low",
        "why": "An economical starting point for extracting dated facts and drafting a short brief from bounded, retrieved sources. Test source accuracy and your review time on a few representative days."
      },
      "escalate": {
        "modelId": "gpt-6.1-sol",
        "when": "Sources conflict, a commitment is ambiguous or review effort stays high. Stronger synthesis can propose an interpretation; priorities and responses still belong to you."
      },
      "alternatives": [
        {
          "modelId": "claude-haiku-5-5",
          "when": "An extraction first alternative in an approved Claude host. Compare the complete workflow, including retrieval, latency, cost and review effort."
        },
        {
          "modelId": "claude-sonnet-5-5",
          "when": "An alternative for deeper synthesis in an approved Claude host when evidence conflicts or the draft requires too much correction."
        }
      ],
      "humanBoundary": "Review inferred commitments, owners and dates against the sources. Choose priorities and how to respond. No model or skill installation grants connector access or authorizes messages, task changes or scheduling."
    },
    "assess-workflow-value": {
      "reviewedOn": "2026-11-02",
      "reviewedLabel": "November 2, 2026",
      "start": {
        "modelId": "gpt-6-luna",
        "effort": "low",
        "why": "A starting point for organizing a small supplied evidence table and explaining deterministic calculations. Check the arithmetic with a calculator or code and review the assumptions."
      },
      "escalate": {
        "modelId": "gpt-6.1-sol",
        "when": "The cases are hard to compare, records conflict or the decision needs a deeper examination. A stronger model cannot recover unmeasured effort."
      },
      "alternatives": [
        {
          "modelId": "claude-haiku-5-5",
          "when": "An alternative for structuring supplied evidence in an approved Claude host. Compare review effort and complete workflow cost on your own cases."
        }
      ],
      "humanBoundary": "The owner decides whether to keep, revise or stop. A model cannot establish savings from missing measurements or infer that released effort produced customer value or growth."
    }
  },
  "catalogUrl": "https://jasminejames.dev/leading-deeper/model-guidance.json"
}
