specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Not Diamond
providerId: notdiamond
created: '2026-06-20'
modified: '2026-06-20'
reconciled: false
tags:
- AI
- LLM
- Model Routing
- Router
- Orchestration
- FinOps
- Cost Management
- FOCUS
description: >-
  FinOps view of Not Diamond spend. Not Diamond charges a small fixed routing fee
  per million tokens routed and positions itself as a cost-optimization layer that
  reduces total LLM spend (cited 20-40% savings) by sending each prompt to the
  cheapest model that still meets quality targets. Underlying model inference is
  billed by the selected providers, not by Not Diamond. Exact routing fee amounts
  are not publicly listed and are left unreconciled.
notes: >-
  Routing fee per 1M tokens is not publicly enumerated; verify against the Not
  Diamond pricing page or a sales quote during reconciliation. Provider inference
  costs are separate and billed by each underlying LLM provider.
sources:
- https://www.notdiamond.ai/pricing
- https://docs.notdiamond.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: Not Diamond
serviceCategory: AI and Machine Learning
billingModel:
  pricingCategory: Usage-Based
  billingFrequency: Monthly
  billingCurrency: USD
  chargeCategories:
  - Usage
  - Purchase
focusColumns:
  ServiceName: Not Diamond Model Router
  ServiceCategory: AI and Machine Learning
  ProviderName: Not Diamond
  PublisherName: Not Diamond
  InvoiceIssuerName: Not Diamond
  BillingCurrency: USD
  ChargeCategory: Usage
  PricingCategory: Usage-Based
meters:
- name: routed_tokens
  description: Tokens routed through the model router, billed as a fixed fee per 1M tokens.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - preference_id
- name: routing_decisions
  description: Number of modelSelect routing decisions made.
  unit: requests
  aggregation: sum
  dimensions:
  - account
  - preference_id
- name: provider_inference_cost
  description: >-
    Inference cost incurred at the selected underlying LLM provider. Billed by the
    provider, not by Not Diamond; tracked here for total cost of ownership.
  unit: usd
  aggregation: sum
  dimensions:
  - account
  - provider
  - model
principles:
- name: Visibility
  description: Use the Not Diamond savings/usage dashboard to see routing decisions and realized savings vs. a single-model baseline.
- name: Allocation
  description: Tag API keys and preference IDs per workload/team; map routed-token spend and downstream provider spend to cost centers.
- name: Optimization
  description: Apply cost or cost_quality_tradeoff settings so the router favors cheaper models when quality permits; train custom routers on real evaluation data to improve savings.
- name: Accountability
  description: Review routed-token fees and underlying provider inference costs monthly against budget per project/key.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com