specification: FinOps Framework
specificationVersion: '1.0'
schema: https://www.finops.org/framework/
provider: Beam
providerId: beam-cloud
created: '2026-06-20'
modified: '2026-06-20'
reconciled: true
tags:
- Serverless
- GPU
- Python
- Inference
- Containers
- FinOps
- Cost Management
- FOCUS
description: >-
  FinOps view of Beam (beam.cloud) spend. Beam bills serverless and sandbox compute
  per second - CPU per core-second, RAM per GB-second, and serverless GPUs per
  GPU-second - while dedicated on-demand GPU containers are metered per hour. Usage
  tiers (Developer, Team, Growth) add a subscription line and $30 of monthly credits;
  the dominant cost driver is per-second GPU container runtime, so cold-start speed,
  autoscaling, and idle keep-alive directly shape the bill.
notes: >-
  Per-second and per-hour rates are subject to change and vary by GPU type and
  region; verify against https://www.beam.cloud/pricing during reconciliation.
sources:
- https://www.beam.cloud/pricing
- https://docs.beam.cloud
- 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: Beam
serviceCategory: Compute
billingModel:
  pricingCategory: Usage-Based
  billingFrequency: Monthly
  billingCurrency: USD
  chargeCategories:
  - Usage
  - Purchase
  - Adjustment
focusColumns:
  ServiceName: Beam
  ServiceCategory: Compute
  ProviderName: Beam
  PublisherName: Beam
  InvoiceIssuerName: Beam
  BillingCurrency: USD
  ChargeCategory: Usage
  PricingCategory: Usage-Based
meters:
- name: cpu_core_seconds
  description: CPU core-seconds consumed by serverless functions, endpoints, task queues, and sandboxes, billed per core per second.
  unit: core_seconds
  aggregation: sum
  dimensions:
  - account
  - app
  - container
- name: ram_gb_seconds
  description: Memory GB-seconds consumed by running containers, billed per GB per second.
  unit: gb_seconds
  aggregation: sum
  dimensions:
  - account
  - app
  - container
- name: gpu_seconds_serverless
  description: Serverless GPU-seconds (e.g. RTX 4090, A10G) consumed by autoscaled containers, billed per second.
  unit: gpu_seconds
  aggregation: sum
  dimensions:
  - account
  - app
  - gpu_type
- name: gpu_hours_on_demand
  description: Dedicated on-demand GPU container hours (e.g. H100, A100, B200), billed per hour.
  unit: hours
  aggregation: sum
  dimensions:
  - account
  - app
  - gpu_type
- name: subscription_fee
  description: Monthly usage-tier subscription (Developer $0, Team $89, Growth custom), inclusive of $30 monthly credits.
  unit: months
  aggregation: sum
  dimensions:
  - account
  - tier
- name: seats
  description: Additional team seats beyond those included in the tier, billed per seat per month.
  unit: seats
  aggregation: sum
  dimensions:
  - account
  - tier
principles:
- name: Visibility
  description: Track per-second CPU/RAM/GPU burn per app and container; watch GPU-seconds as the primary cost driver.
- name: Allocation
  description: Tag apps/deployments per workload or team and map container usage to cost centers.
- name: Optimization
  description: Use fast cold starts and tight autoscaling to avoid idle GPU-seconds; right-size GPU type per workload; move long jobs to task queues and prefer cheaper GPUs (RTX 4090/A10G) when sufficient.
- name: Accountability
  description: Assign owners per app/key; review per-second compute spend monthly against the $30 credit and tier budget.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com