specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: Beam
providerId: beam-cloud
created: '2026-06-20'
modified: '2026-06-20'
reconciled: true
tags:
- Serverless
- GPU
- Python
- Inference
- Containers
- Plans
description: >-
  Beam combines usage-tier subscriptions with pay-per-second compute. Serverless
  and sandbox workloads bill by the second for CPU cores, RAM, and attached GPUs,
  while dedicated on-demand GPU containers are quoted per hour. Every tier includes
  $30 in monthly credits, unlimited apps, storage volumes, custom images, and API
  requests; tiers differ mainly by GPU/CPU container concurrency, seats, and log
  retention.
notes: >-
  Per-second and per-hour rates reflect the Beam pricing page as of 2026-06-20 and
  start-from quotes that vary by region and availability; verify the current full
  GPU catalog and rates at https://www.beam.cloud/pricing during reconciliation.
sources:
- https://www.beam.cloud/pricing
- https://docs.beam.cloud
plans:
- id: beam-developer
  name: Developer
  type: subscription
  description: >-
    Free tier with $30 monthly credits, 1 seat, up to 5 GPU containers and 30 CPU
    containers of concurrency, and 30-day log retention.
  entries:
  - label: Monthly Subscription
    name: developer_subscription
    type: flat
    metric: subscription
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $0/mo (includes $30 monthly credits)
    userMultiplied: false
  - label: GPU Container Concurrency
    name: developer_gpu_concurrency
    type: limit
    metric: containers
    limit: 5
    timeFrame: month
    geo: global
    unit: 1
    price: included
    userMultiplied: false
  - label: CPU Container Concurrency
    name: developer_cpu_concurrency
    type: limit
    metric: containers
    limit: 30
    timeFrame: month
    geo: global
    unit: 1
    price: included
    userMultiplied: false
  elements:
  - name: Web Endpoints
  - name: Task Queues
  - name: Sandboxes
  - name: Storage Volumes
- id: beam-team
  name: Team
  type: subscription
  description: >-
    $89/mo with $30 monthly credits, 3 seats included (+$25 each), up to 50 GPU
    containers and 1,000 CPU containers of concurrency, and 30-day log retention.
  entries:
  - label: Monthly Subscription
    name: team_subscription
    type: flat
    metric: subscription
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $89/mo (includes $30 monthly credits)
    userMultiplied: false
  - label: Additional Seat
    name: team_seat
    type: flat
    metric: seats
    limit: -1
    timeFrame: month
    geo: global
    unit: 1
    price: $25/seat beyond 3 included
    userMultiplied: true
  - label: GPU Container Concurrency
    name: team_gpu_concurrency
    type: limit
    metric: containers
    limit: 50
    timeFrame: month
    geo: global
    unit: 1
    price: included
    userMultiplied: false
  - label: CPU Container Concurrency
    name: team_cpu_concurrency
    type: limit
    metric: containers
    limit: 1000
    timeFrame: month
    geo: global
    unit: 1
    price: included
    userMultiplied: false
  elements:
  - name: Web Endpoints
  - name: Task Queues
  - name: Sandboxes
  - name: Storage Volumes
- id: beam-growth
  name: Growth
  type: enterprise
  description: >-
    Custom contract with $30 monthly credits, unlimited seats, custom GPU
    concurrency, unlimited CPU concurrency, and 1-year log retention. Contact Beam.
  entries:
  - label: Enterprise Agreement
    name: growth_contract
    type: flat
    metric: contract
    limit: -1
    timeFrame: year
    geo: global
    unit: 1
    price: contact sales
    userMultiplied: false
  elements:
  - name: Custom GPU Concurrency
  - name: Unlimited CPU Concurrency
  - name: 1-Year Log Retention
- id: beam-serverless-compute
  name: Serverless and Sandbox Compute
  type: usage
  description: >-
    Pay-per-second metering for serverless functions, endpoints, task queues, and
    sandboxes. CPU and RAM are billed per core-second and per GB-second; serverless
    GPUs are billed per second.
  entries:
  - label: CPU
    name: cpu_core_second
    type: usage
    metric: core_seconds
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: $0.0000528 per core per second
    userMultiplied: false
  - label: RAM
    name: ram_gb_second
    type: usage
    metric: gb_seconds
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: $0.0000056 per GB per second
    userMultiplied: false
  - label: RTX 4090 (serverless)
    name: gpu_rtx4090_second
    type: usage
    metric: gpu_seconds
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: $0.000192 per second
    userMultiplied: false
  - label: A10G (serverless)
    name: gpu_a10g_second
    type: usage
    metric: gpu_seconds
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: $0.000292 per second
    userMultiplied: false
  elements:
  - name: Web Endpoints
  - name: Task Queues
  - name: Sandboxes
- id: beam-on-demand-gpu
  name: On-Demand GPU Containers
  type: usage
  description: >-
    Dedicated on-demand GPU containers billed per hour, starting-from rates by GPU
    type.
  entries:
  - label: B200 SXM6
    name: gpu_b200
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $3.93/hr
    userMultiplied: false
  - label: H200 SXM5
    name: gpu_h200
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $1.99/hr
    userMultiplied: false
  - label: H100 PCIe
    name: gpu_h100
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $1.74/hr
    userMultiplied: false
  - label: A100 80GB SXM4
    name: gpu_a100_80gb
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $1.30/hr
    userMultiplied: false
  - label: RTX PRO 6000 PCIe
    name: gpu_rtx_pro_6000
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $1.04/hr
    userMultiplied: false
  - label: L40S PCIe
    name: gpu_l40s
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $0.72/hr
    userMultiplied: false
  - label: RTX 5090 PCIe
    name: gpu_rtx5090
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $0.68/hr
    userMultiplied: false
  - label: A6000 PCIe
    name: gpu_a6000
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $0.51/hr
    userMultiplied: false
  - label: RTX 4090 PCIe
    name: gpu_rtx4090_hr
    type: usage
    metric: hours
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1
    price: from $0.42/hr
    userMultiplied: false
  elements:
  - name: Dedicated GPU Containers
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com