specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Tensorlake
providerId: tensorlake
created: '2026-07-12'
modified: '2026-07-12'
reconciled: false
tags:
- Document Extraction
- Document Ingestion
- OCR
- AI
- FinOps
- Cost Management
- FOCUS
description: >-
  FinOps view of Tensorlake Cloud spend. Tensorlake bills on a credit model
  where 1 credit equals $0.01. Document ingestion is metered primarily by the
  page (published rates put parsing and extraction near $10 per 1,000 pages),
  while serverless workflows and MicroVM sandboxes are metered on compute - CPU
  per hour, RAM per GB-hour, disk per GB-hour, and snapshot storage per
  GB-month - billed by the second. A Pro subscription ($250/month + usage) adds
  discounted compute rates, and Enterprise agreements cover high-volume or
  in-VPC deployments.
notes: >-
  Per-unit rates are not reconciled here. The $10-per-1,000-pages document rate
  and the sandbox compute meters come from the Tensorlake pricing page and
  third-party listings; verify current numbers during reconciliation.
sources:
- https://www.tensorlake.ai/pricing
- https://docs.tensorlake.ai
- https://focus.finops.org/focus-specification/v1-3/
alignedWith:
  framework: FinOps Foundation Framework
  frameworkUrl: https://www.finops.org/framework/
  dataSpec: FOCUS
  dataSpecVersion: '1.3'
  dataSpecUrl: https://focus.finops.org/focus-specification/v1-3/
publisherName: Tensorlake
serviceCategory: AI and Machine Learning
billingModel:
  pricingCategory: Usage-Based
  billingFrequency: Monthly
  billingCurrency: USD
  chargeCategories:
  - Usage
  - Purchase
  - Adjustment
focusColumns:
  ServiceName: Tensorlake Cloud
  ServiceCategory: AI and Machine Learning
  ProviderName: Tensorlake
  PublisherName: Tensorlake
  InvoiceIssuerName: Tensorlake
  BillingCurrency: USD
  ChargeCategory: Usage
  PricingCategory: Usage-Based
meters:
- name: document_pages
  description: Pages of documents parsed / extracted through the Document Ingestion API, billed in credits (~$10 per 1,000 pages).
  unit: pages
  aggregation: sum
  dimensions:
  - project
  - dataset
- name: sandbox_cpu
  description: CPU time consumed by MicroVM sandboxes, billed per hour and metered by the second.
  unit: hours
  aggregation: sum
  dimensions:
  - project
  - sandbox
- name: sandbox_memory
  description: RAM consumed by sandboxes, billed per GB-hour.
  unit: gb_hours
  aggregation: sum
  dimensions:
  - project
  - sandbox
- name: sandbox_disk
  description: Disk consumed by sandboxes, billed per GB-hour.
  unit: gb_hours
  aggregation: sum
  dimensions:
  - project
  - sandbox
- name: snapshot_storage
  description: Sandbox snapshot storage, billed per GB-month.
  unit: gb_months
  aggregation: sum
  dimensions:
  - project
principles:
- name: Visibility
  description: Track pages processed per project and dataset, and sandbox CPU/RAM/disk/snapshot usage per sandbox, to see where credits are spent.
- name: Allocation
  description: Name projects, datasets, and sandboxes per workload/team so page and compute spend map to internal cost centers.
- name: Optimization
  description: Scope page ranges and chunking to only the content you need; suspend idle sandboxes to pause compute billing; consolidate repeated document types into datasets with a shared configuration.
- name: Accountability
  description: Assign owners per project; review monthly page volume and sandbox compute against the plan's credit allowance or Pro subscription.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com