{
  "schema": "varexis-public-evidence-record/0.3",
  "generated_at_utc": "2026-10-24T17:38:21.804Z",
  "record_id": "VXI-EXT-PROCUREMENT-VERCEL-GATEWAY-FALLBACK-MODEL-COST-ATTRIBUTION-01-v1",
  "record_kind": "external_reference",
  "title": "AI Gateway fallback identity and cost attribution",
  "system": "Vercel",
  "evidence_family": "vendor_evidence_review",
  "index_status": "PUBLIC_SAFE_SOURCE_REVIEW",
  "evidence_state": "PRIMARY_DOCUMENTATION_REVIEW_ONLY",
  "canonical_question": "Does using one AI Gateway endpoint mean every fallback request used the same model and cost basis?",
  "near_questions": [
    "Can an AI Gateway fallback change the model used?",
    "Where does Vercel report model and provider fallback attempts?",
    "Can normalized Python gateway metadata omit routing details?",
    "What evidence attributes cost after a model fallback?"
  ],
  "supported_claims": [
    {
      "claim": "Vercel AI Gateway can try different models and providers after a failure, and providers have different pricing characteristics. The modelAttempts metadata identifies attempted models and provider attempts. The guide warns that some normalized Python beta metadata can omit routing details, requiring the raw response or request logs.",
      "reliance": "DOCUMENTED_CAPABILITY_ONLY"
    }
  ],
  "not_demonstrated_claims": [
    "A claim that every failed provider attempt is billable",
    "Which attempts were billed and the final invoice are not established by routing metadata alone.",
    "Buyer deployment testing, live runtime enforcement, certification or permission to act",
    "Expanded capabilities or licences for existing INDEX executables"
  ],
  "public_summary": "Vercel AI Gateway can try different models and providers after a failure, and providers have different pricing characteristics. The modelAttempts metadata identifies attempted models and provider attempts. The guide warns that some normalized Python beta metadata can omit routing details, requiring the raw response or request logs.",
  "related_record_ids": [],
  "source_review": {
    "source_type": "official_external_documentation",
    "source_title": "AI Gateway Model Fallbacks",
    "source_url": "https://vercel.com/docs/ai-gateway/models-and-providers/model-fallbacks",
    "checked_on": "2026-10-24",
    "candidate_record_id": "VXI-EXT-PROCUREMENT-VERCEL-GATEWAY-FALLBACK-MODEL-COST-ATTRIBUTION-01-v1",
    "candidate_record_sha256": "e699dbff1bea53f5ac0507776f0c0c7f1eeca1d460c3a3865504450481bef19d",
    "method": "Primary documentation reviewed for the published bounded statement.",
    "deployment_tested_by_varexis": false,
    "revalidation": "The delivered snapshot preserves the review date shown. Recheck current primary documentation and deployment context before relying on present applicability; payment does not trigger fresh research."
  },
  "answer_policy": {
    "classification": "DOCUMENTED_WITH_DEPLOYMENT_BOUNDARY",
    "bounded_answer": "Vercel AI Gateway can try different models and providers after a failure, and providers have different pricing characteristics. The modelAttempts metadata identifies attempted models and provider attempts. The guide warns that some normalized Python beta metadata can omit routing details, requiring the raw response or request logs.",
    "required_buyer_context": "Requested and actual model/provider identifiers; Fallback attempt metadata or raw request logs; Usage and applicable billing records. Buyer deployment not inspected.",
    "missing_evidence": "A broader claim needs evidence matching the intended use. Payment does not upgrade the evidence state."
  },
  "private_boundary": "Raw logs, thresholds, private paths, source code, protected mechanisms and reconstruction-level details are excluded from public and paid delivery.",
  "decision_profile": {
    "schema": "varexis-vendor-evidence-review/1",
    "review_status": "documentation_review_only",
    "checked_on": "2026-10-24",
    "claim_to_check": "Does using one AI Gateway endpoint mean every fallback request used the same model and cost basis?",
    "documented_statement": {
      "status": "documented",
      "statement": "Vercel AI Gateway can try different models and providers after a failure, and providers have different pricing characteristics. The modelAttempts metadata identifies attempted models and provider attempts. The guide warns that some normalized Python beta metadata can omit routing details, requiring the raw response or request logs.",
      "source_ids": [
        "vercel-fallback",
        "vercel-metadata"
      ]
    },
    "review_inference": {
      "status": "review_inference",
      "statement": "Reconcile the served model/provider and measured usage before assigning request cost.",
      "source_ids": [
        "vercel-fallback",
        "vercel-metadata"
      ]
    },
    "buyer_evidence_required": [
      "Requested and actual model/provider identifiers",
      "Fallback attempt metadata or raw request logs",
      "Usage and applicable billing records"
    ],
    "review_condition": {
      "status": "varexis_review_policy",
      "statement": "Do not infer the served model or cost basis from the shared endpoint alone."
    },
    "unknowns": "Which attempts were billed and the final invoice are not established by routing metadata alone.",
    "unknown_finding": {
      "status": "unknown",
      "statement": "Which attempts were billed and the final invoice are not established by routing metadata alone."
    },
    "evidence_sources": [
      {
        "source_id": "vercel-fallback",
        "url": "https://vercel.com/docs/ai-gateway/models-and-providers/model-fallbacks",
        "final_url": "https://vercel.com/docs/ai-gateway/models-and-providers/model-fallbacks",
        "status": 200,
        "captured_at_utc": "2026-10-24T17:28:00.811Z",
        "sha256": "73d56ec03886c1a4c166bc66a701588b40a0ee07e8a29f495ce610aae9f65edd",
        "bytes": 1216243,
        "title": "AI Gateway Model Fallbacks",
        "hash_basis": "Decoded HTTP response body bytes",
        "rights_review": "Factual paraphrase and source metadata only; raw source retained privately and not redistributed.",
        "checked_on": "2026-10-24",
        "capture_status": 200,
        "claim_review_completed": true,
        "reviewed_sections": [
          "Model fallbacks",
          "modelAttempts",
          "Normalized Python beta metadata limitation"
        ],
        "version_context": "Rolling official documentation captured 2026-10-24; current model/API version, account settings and applicable billing or retention terms require verification."
      },
      {
        "source_id": "vercel-metadata",
        "url": "https://vercel.com/docs/ai-gateway/models-and-providers/provider-options",
        "final_url": "https://vercel.com/docs/ai-gateway/models-and-providers/provider-options",
        "status": 200,
        "captured_at_utc": "2026-10-24T17:28:01.216Z",
        "sha256": "f9104b24d3fedd844ca0a5d87fbb32c4e0d0b25c8dd338b113709f3d28048f4c",
        "bytes": 1366439,
        "title": "AI Gateway Provider Routing and Fallbacks",
        "hash_basis": "Decoded HTTP response body bytes",
        "rights_review": "Factual paraphrase and source metadata only; raw source retained privately and not redistributed.",
        "checked_on": "2026-10-24",
        "capture_status": 200,
        "claim_review_completed": true,
        "reviewed_sections": [
          "Provider routing",
          "Pricing differences"
        ],
        "version_context": "Rolling official documentation captured 2026-10-24; current model/API version, account settings and applicable billing or retention terms require verification."
      }
    ],
    "observed_deployment": false,
    "commercial_reliance_boundary": "Documentation review only; the buyer must establish applicability and actual outcomes. INDEX does not inspect or control the deployment.",
    "paid_value": {
      "decision_context": "Prevent a successful fallback from being costed against the wrong model.",
      "free": "Complete review, sources, practical questions and limitations; may be saved locally.",
      "single_request": "Existing server-saved version-bound route snapshot for later authenticated retrieval.",
      "evidence_receipt": "The same snapshot plus the existing unsigned delivery receipt.",
      "excluded": [
        "Fresh vendor research",
        "Continuous monitoring or change alerts",
        "Testing the buyer configuration",
        "Private evidence or vendor certification"
      ],
      "integrity": "Hashes identify content versions; they do not prove truth or provide a signature."
    },
    "revalidation": {
      "trigger": "A saved purchase does not recheck the source or buyer configuration.",
      "specific_watch": "Recheck the documented behavior, product version and buyer evidence before relying on a current outcome."
    },
    "buyer_relevance": "Prevent a successful fallback from being costed against the wrong model.",
    "refusal_conditions": [
      "Do not infer the served model or cost basis from the shared endpoint alone."
    ]
  },
  "niche": {
    "niche_id": "agent-authority-action-evidence",
    "label": "Agent Authority & Action Evidence",
    "classification_id": "payments",
    "classification_label": "Agent Payments, Authority & Transaction Evidence",
    "topic_id": "payments",
    "buyer_intent": "Prevent a successful fallback from being costed against the wrong model.",
    "comparison_role": "supplementary_discovery"
  },
  "paid_access_note": "The full published answer and boundaries are free. Paid access saves the selected route and source versions, with an unsigned receipt when purchased. Preview coverage before paying. Fresh research, bespoke assessment and protected evidence are excluded.",
  "service_route": {
    "request_guide": "/varexis-index/request.html",
    "request_workflow": "/varexis-index/request-workflow.json",
    "service_id": "single_request",
    "service_routes_url": "/varexis-index/service-routes.json",
    "payment_required": true,
    "availability_endpoint": "/api/index/capabilities",
    "operational_status": "AUTOMATIC_SERVICE_CAPABILITIES_GATED",
    "checkout_url": "/index-access"
  },
  "authority_boundary": "VAREXIS-INDEX informs evidence reliance. It does not provide advice, certification, deployment authority, security assurance, production approval, or guarantee.",
  "human_page": "/varexis-index/catalog.html#vercel-gateway-fallback-model-cost-attribution",
  "machine_readable_record": "/varexis-index/records/vercel-gateway-fallback-model-cost-attribution.json"
}
