Beam · Pricing Plans

Beam Cloud Plans Pricing

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.

Beam Cloud Plans Pricing is the machine-readable pricing-plan profile for Beam on the APIs.io network, conforming to the API Commons Plans specification.

It defines 5 plans, covering subscription, enterprise, and usage tiers, with named plans including Developer, Team, Growth, Serverless and Sandbox Compute, On-Demand GPU Containers.

Tagged areas include Serverless, GPU, Python, Inference, and Containers.

5 Plans API Commons Plans
View Source
ServerlessGPUPythonInferenceContainersPlans

Plans

Developer subscription

Free tier with $30 monthly credits, 1 seat, up to 5 GPU containers and 30 CPU containers of concurrency, and 30-day log retention.

Monthly Subscription (subscription · month) $0/mo (includes $30 monthly credits) USD
GPU Container Concurrency (containers · month) included USD
CPU Container Concurrency (containers · month) included USD
Team subscription

$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.

Monthly Subscription (subscription · month) $89/mo (includes $30 monthly credits) USD
Additional Seat (seats · month) $25/seat beyond 3 included USD
GPU Container Concurrency (containers · month) included USD
CPU Container Concurrency (containers · month) included USD
Growth enterprise

Custom contract with $30 monthly credits, unlimited seats, custom GPU concurrency, unlimited CPU concurrency, and 1-year log retention. Contact Beam.

Enterprise Agreement (contract · year) contact sales USD
Serverless and Sandbox Compute usage

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.

CPU (core_seconds · usage) $0.0000528 per core per second USD
RAM (gb_seconds · usage) $0.0000056 per GB per second USD
RTX 4090 (serverless) (gpu_seconds · usage) $0.000192 per second USD
A10G (serverless) (gpu_seconds · usage) $0.000292 per second USD
On-Demand GPU Containers usage

Dedicated on-demand GPU containers billed per hour, starting-from rates by GPU type.

B200 SXM6 (hours · usage) from $3.93/hr USD
H200 SXM5 (hours · usage) from $1.99/hr USD
H100 PCIe (hours · usage) from $1.74/hr USD
A100 80GB SXM4 (hours · usage) from $1.30/hr USD
RTX PRO 6000 PCIe (hours · usage) from $1.04/hr USD
L40S PCIe (hours · usage) from $0.72/hr USD
RTX 5090 PCIe (hours · usage) from $0.68/hr USD
A6000 PCIe (hours · usage) from $0.51/hr USD
RTX 4090 PCIe (hours · usage) from $0.42/hr USD

Sources