specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: DataCrunch
providerId: datacrunch
created: '2026-06-21'
modified: '2026-06-21'
reconciled: true
tags:
- GPU Cloud
- Infrastructure
- Compute
- Inference
- Serverless
- Europe
- Plans
description: >-
  DataCrunch uses transparent per-GPU-per-hour, pay-as-you-go pricing for
  on-demand GPU instances, with lower long-term (reserved) rates available via
  discount-period contracts. Serverless containers are billed per second of
  active compute on a chosen container compute size with autoscaling, so you
  pay only while replicas are running. Block storage volumes are billed per GB
  per month. There are no fixed Free / Developer / Enterprise software tiers;
  customers add credit (balance) and pay for the infrastructure they consume.
notes: >-
  Representative on-demand per-GPU-hour rates from the DataCrunch pricing page
  (provider is rebranding to Verda; api.datacrunch.io remains the API host).
  Rates change frequently and reserved/long-term contracts are cheaper; verify
  current rates on the pricing page.
sources:
- https://datacrunch.io/pricing
- https://datacrunch.io/products
- https://docs.datacrunch.io/
plans:
- id: datacrunch-on-demand
  name: On-Demand GPU Instances
  type: usage
  description: >-
    Per-GPU-per-hour on-demand pricing for dedicated GPU instances with no
    commitment. Billed by the second while the instance runs.
  entries:
  - label: NVIDIA B200 (192GB)
    name: b200
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $3.99 per GPU/hour
    userMultiplied: false
  - label: NVIDIA H200 SXM (141GB)
    name: h200
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $2.99 per GPU/hour
    userMultiplied: false
  - label: NVIDIA H100 SXM (80GB)
    name: h100
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $2.29 per GPU/hour
    userMultiplied: false
  - label: NVIDIA A100 SXM (80GB)
    name: a100_80gb
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $1.29 per GPU/hour
    userMultiplied: false
  - label: NVIDIA A100 (40GB)
    name: a100_40gb
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $0.72 per GPU/hour
    userMultiplied: false
  - label: NVIDIA L40S (48GB)
    name: l40s
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $0.91 per GPU/hour
    userMultiplied: false
  - label: NVIDIA RTX 6000 Ada (48GB)
    name: rtx6000ada
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $0.83 per GPU/hour
    userMultiplied: false
  - label: NVIDIA RTX A6000 (48GB)
    name: rtxa6000
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $0.49 per GPU/hour
    userMultiplied: false
  - label: NVIDIA V100 (32GB)
    name: v100
    type: usage
    metric: gpu_hours
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: $0.14 per GPU/hour
    userMultiplied: false
  elements:
  - name: GPU Instances
  - name: Instance Types & Availability
  - name: Images & Startup Scripts
  - name: SSH Keys
- id: datacrunch-long-term
  name: Long-Term / Reserved Contracts
  type: subscription
  description: >-
    Discounted per-GPU-hour rates for committed long-term contracts over
    defined discount periods (e.g. monthly, multi-month), available for both
    instances and clusters.
  entries:
  - label: Reserved GPU Instance
    name: reserved_instance
    type: subscription
    metric: gpu_hours
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: discounted vs. on-demand (see long-term periods)
    userMultiplied: false
  elements:
  - name: Long-Term Periods (Instances)
  - name: Long-Term Periods (Clusters)
- id: datacrunch-serverless-containers
  name: Serverless Containers
  type: usage
  description: >-
    Autoscaling serverless container deployments billed per second of active
    compute on the selected container compute size; replicas scale to zero so
    you pay only while serving.
  entries:
  - label: Container Compute
    name: container_compute
    type: usage
    metric: compute_seconds
    limit: -1
    timeFrame: hour
    geo: global
    unit: 1
    price: per-second billing by container compute type (see container-types)
    userMultiplied: false
  elements:
  - name: Serverless Containers
  - name: OpenAI-Compatible Inference
- id: datacrunch-storage
  name: Block Storage Volumes
  type: usage
  description: >-
    Persistent NVMe/HDD block storage volumes billed per GB per month,
    independent of instance runtime.
  entries:
  - label: Volume Storage
    name: volume_storage
    type: usage
    metric: gb
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: per GB/month (see volume-types)
    userMultiplied: false
  elements:
  - name: Volumes
  - name: Volume Types
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com