specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Groq
providerId: groq
created: '2026-05-08'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-08, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-08'
reconciled: true
tags:
- AI
- LLM
- Inference
- LPU
- Low Latency
- FinOps
- Cost Management
- FOCUS
description: >-
  FinOps view of GroqCloud spend. Groq bills usage-based per-token rates for
  chat / vision / reasoning per model, per-million-character rates for TTS,
  per-hour transcription rates for STT, per-call or per-hour rates for tools,
  and a 50% Batch discount. Prompt Caching gives 50% off cached input tokens.
notes: >-
  Per-model rates are subject to change; verify against the Groq pricing page
  during reconciliation.
sources:
- https://groq.com/pricing
- https://console.groq.com/docs
- 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: Groq
serviceCategory: AI and Machine Learning
billingModel:
  pricingCategory: Usage-Based
  billingFrequency: Monthly
  billingCurrency: USD
  chargeCategories:
  - Usage
  - Purchase
  - Adjustment
focusColumns:
  ServiceName: GroqCloud
  ServiceCategory: AI and Machine Learning
  ProviderName: Groq
  PublisherName: Groq
  InvoiceIssuerName: Groq
  BillingCurrency: USD
  ChargeCategory: Usage
  PricingCategory: Usage-Based
meters:
- name: input_tokens
  description: Tokens sent in chat / vision / reasoning requests, billed per 1M tokens per model.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
  - api
- name: cached_input_tokens
  description: Cached-input tokens billed at 50% of the standard input rate.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
- name: output_tokens
  description: Tokens generated, billed per 1M tokens per model.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
  - api
- name: tts_characters
  description: TTS characters synthesized, billed per 1M characters per voice/model.
  unit: characters
  aggregation: sum
  dimensions:
  - account
  - model
- name: stt_audio_hours
  description: Audio hours transcribed, billed per hour per Whisper variant.
  unit: hours
  aggregation: sum
  dimensions:
  - account
  - model
- name: tool_invocations
  description: Tool calls (web search, Wolfram) priced per 1,000 invocations.
  unit: invocations
  aggregation: sum
  dimensions:
  - account
  - tool
- name: tool_compute_hours
  description: Tool compute hours (e.g., Code Execution at $0.18/hr).
  unit: hours
  aggregation: sum
  dimensions:
  - account
  - tool
- name: batch_tokens
  description: Tokens consumed via the Batch API at 50% discount.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
- name: flex_tokens
  description: Tokens consumed via Flex Processing tier at relaxed-latency discount.
  unit: tokens
  aggregation: sum
  dimensions:
  - account
  - model
principles:
- name: Visibility
  description: Pull GroqCloud usage and billing exports; inspect per-model token burn.
- name: Allocation
  description: Tag API keys per workload/team; map to internal cost centers.
- name: Optimization
  description: Use Batch (-50%) for non-realtime jobs; enable Prompt Caching; route lower-quality work to Flex; pick smaller open-source models when sufficient.
- name: Accountability
  description: Assign owners per project/key; review token spend monthly against budget.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com