specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Featherless AI
providerId: featherless
created: '2026-06-21'
modified: '2026-06-21'
reconciled: true
tags:
- AI
- LLM
- Inference
- Serverless
- Open Models
- FinOps
- Cost Management
- FOCUS
description: >-
  FinOps view of Featherless AI spend. Featherless bills a flat monthly
  subscription per plan (and per unit for business tiers) with unlimited
  tokens, rather than usage-based per-token rates. Cost is therefore driven
  by the number and tier of active subscriptions and the concurrency units
  purchased, not by token volume. Optimization focuses on right-sizing the
  plan tier (model-size and concurrency needs) instead of reducing token burn.
notes: >-
  Subscription prices and tier limits are subject to change; verify against
  the Featherless plans page during reconciliation.
sources:
- https://featherless.ai/pricing
- https://featherless.ai/docs/plans
- 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: Featherless AI
serviceCategory: AI and Machine Learning
billingModel:
  pricingCategory: Subscription-Based
  billingFrequency: Monthly
  billingCurrency: USD
  chargeCategories:
  - Purchase
  - Usage
  - Adjustment
focusColumns:
  ServiceName: Featherless AI
  ServiceCategory: AI and Machine Learning
  ProviderName: Featherless AI
  PublisherName: Featherless AI
  InvoiceIssuerName: Featherless AI
  BillingCurrency: USD
  ChargeCategory: Purchase
  PricingCategory: Subscription-Based
meters:
- name: subscription_plan
  description: Flat monthly subscription per plan tier (Basic, Premium, agent, business).
  unit: subscription
  aggregation: sum
  dimensions:
  - account
  - plan
- name: concurrency_units
  description: Concurrency units purchased on per-unit business plans.
  unit: units
  aggregation: sum
  dimensions:
  - account
  - plan
- name: concurrent_connections
  description: Concurrent in-flight requests consumed against the plan concurrency limit.
  unit: connections
  aggregation: max
  dimensions:
  - account
  - plan
- name: tokens
  description: Tokens processed; unlimited and not billed under the subscription model. Tracked for utilization only.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
principles:
- name: Visibility
  description: Track active subscriptions, plan tiers, and concurrency-unit counts; monitor live concurrency utilization via the account concurrency stream.
- name: Allocation
  description: Map each subscription/API key to a workload or team and assign plan cost to cost centers.
- name: Optimization
  description: Right-size the plan tier - downgrade when model-size and concurrency needs are lower, consolidate workloads to fully utilize unlimited-token concurrency before buying more units.
- name: Accountability
  description: Assign owners per subscription; review whether each active plan tier and unit count is justified by utilization monthly.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com