specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: CentML
providerId: centml
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- AI
- LLM
- Inference
- Serverless
- GPU
- Plans
description: >-
  CentML uses a credit-based, pay-as-you-go pricing model where 1 CentML credit
  equals 1 USD. Serverless endpoints are billed by the total number of input and
  output tokens processed, with per-token rates that vary by model. Dedicated
  deployments (dedicated inference / compute endpoints) are billed by the type
  and duration of GPU hardware used, on a per-minute basis (commonly expressed as
  a per-GPU-hour rate). Enterprise plans with custom commitments are available by
  contacting CentML. Specific per-model token rates and per-GPU-hour rates are not
  reconciled in this artifact and should be confirmed on the CentML pricing page.
notes: >-
  Representative billing structure as published by CentML; exact per-model
  serverless token rates and per-GPU-hour dedicated rates change over time and
  are not reconciled here. Verify on the CentML pricing page during reconciliation.
sources:
- https://centml.ai/pricing/
- https://docs.centml.ai/apps/serverless
- https://docs.centml.ai/apps/inference
plans:
- id: centml-serverless
  name: Serverless Endpoints
  type: usage
  description: >-
    OpenAI-compatible serverless inference billed by tokens processed. No
    infrastructure to manage; per-token rates vary by model.
  entries:
  - label: Serverless Input Tokens
    name: serverless_input_tokens
    type: usage
    metric: tokens
    limit: -1
    timeFrame: month
    geo: global
    unit: 1000000
    price: per-1M-token rate varies by model (see CentML pricing page)
    userMultiplied: false
  - label: Serverless Output Tokens
    name: serverless_output_tokens
    type: usage
    metric: tokens
    limit: -1
    timeFrame: month
    geo: global
    unit: 1000000
    price: per-1M-token rate varies by model (see CentML pricing page)
    userMultiplied: false
  elements:
  - name: Chat Completions
  - name: Completions
  - name: Models
- id: centml-dedicated
  name: Dedicated Deployments
  type: usage
  description: >-
    Dedicated, autoscaling inference / compute endpoints billed by GPU hardware
    type and duration on a per-minute basis (per-GPU-hour equivalent).
  entries:
  - label: Dedicated GPU Hours
    name: dedicated_gpu_hours
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: per-GPU-hour rate varies by hardware instance (A100, H100, L4, etc.)
    userMultiplied: false
  elements:
  - name: Dedicated Inference Endpoints
  - name: Compute Deployments
  - name: Custom Model Endpoints
- id: centml-enterprise
  name: Enterprise
  type: enterprise
  description: >-
    Custom plans for larger-scale or specialized deployments, including volume
    commitments and dedicated support. Contact CentML sales.
  entries:
  - label: Enterprise Agreement
    name: enterprise
    type: flat
    metric: contract
    limit: -1
    timeFrame: year
    geo: global
    unit: 1
    price: contact sales
    userMultiplied: false
  elements:
  - name: Custom Volume Pricing
  - name: Dedicated Support
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com