specification: FinOps Framework specificationVersion: '1.0' schema: https://www.finops.org/framework/ generated: '2026-07-18' method: searched provider: Arkham providerId: arkham tags: - FinOps - Cost Management - Blockchain Intelligence description: >- Arkham API billing is a usage-based credit model. Every request consumes credits from a monthly allowance; cost is per-call (credits = endpoint weight) or per-row (credits = weight x rows returned). Failed 4xx/5xx requests are not charged. The x402 surface prices the same credit weights at $0.20 per credit in USDC on Base. sources: - https://arkm.com/llms/guides/credit-pricing.md - https://arkm.com/llms/guides/rate-limits.md - https://arkm.com/llms/guides/x402-pay-per-request-for-agents.md publisherName: Arkham Intelligence serviceCategory: Blockchain Intelligence billingModel: type: usage-based-credits unit: credit monetary_rate: $0.20 per credit (x402) charge_on_failure: false metering: per_call: credits = endpoint weight per_row: credits = weight x rows returned credit_weights_examples: - {endpoint: 'GET /intelligence/address/{address}', credits: 1, type: per-call} - {endpoint: 'GET /intelligence/address/{address}/all', credits: 2, type: per-call} - {endpoint: 'GET /intelligence/address_enriched/{address}/all', credits: 4, type: per-call} - {endpoint: 'POST /intelligence/address/batch', credits: 250, type: per-call} - {endpoint: 'POST /intelligence/address_enriched/batch/all', credits: 1000, type: per-call} - {endpoint: 'GET /intelligence/search', credits: 30, type: per-call} - {endpoint: 'GET /transfers', credits: 2, type: per-row} - {endpoint: 'GET /transfers/unenriched', credits: 1, type: per-row} - {endpoint: 'GET /swaps', credits: 2, type: per-row} - {endpoint: 'GET /token/top', credits: 10, type: per-call} - {endpoint: 'GET /token/holders/{id}', credits: 30, type: per-call} - {endpoint: 'POST /ws/sessions', credits: 500, type: per-call} - {endpoint: 'GET /ws/transfers', credits: 2, type: per-row} cost_controls: - Use the limit parameter to cap per-row cost. - Cache responses; paginate correctly; use batch/bulk endpoints. - Repeated lookups of the same labeled address within a period count once (label limits).