Skip to content
← Tools / Skills
ToolBeam/10 trustVerified
Beam logo

Beam

Serverless GPU cloud for AI workloads, powered by the open-source Beta9 engine

Visit official site →

About

Beam is a serverless GPU platform for AI workloads: a Pythonic SDK to deploy autoscaling inference endpoints, run background task queues, schedule jobs and spin up isolated Sandboxes for agent code execution — on Beam's cloud GPUs or on your own hardware pooled into the platform. The underlying engine, Beta9, is open source (AGPL-3.0) and self-hostable.

Note: beam-cloud/beam is not a real repository — the code lives at beam-cloud/beta9 (the product's former name). Independent coverage of the company is thinner than for E2B or Modal, and funding figures floating on aggregators are unverified.

What it can do

  • • `@endpoint` decorator: autoscaling inference APIs with GPU, cpu and memory declared in code
  • • Task queues for background jobs (Celery-style); scheduled jobs and webhooks
  • • Sandbox API for isolated agent code execution
  • • Volumes for distributed storage; bring-your-own-GPU pools joinable to the platform
  • • Fast cold starts via custom container runtime and embedded caching (provider-claimed)
  • • Beta9 engine is AGPL-3.0 open source and self-hostable

Setup

  1. Create an account at https://beam.cloud.
  2. Create a venv and install the client: python3 -m venv .venv && source .venv/bin/activate && python -m pip install --upgrade beam-client.
  3. Sign in: beam login (finishes in the browser); verify with beam whoami.
  4. Write app.py with an @endpoint-decorated function (GPU, cpu and memory declared in the decorator).
  5. Deploy: beam deploy app.py:predict; call the printed URL with a token from platform.beam.cloud → Settings → API Keys (beam logs <name> --follow streams logs).

Prerequisites

Beam account + browser-based `beam login`; Python 3.x virtualenv; BEAM_TOKEN from platform.beam.cloud API keys to call deployed endpoints.

Use cases

codeautomationdatamedia

Works with

Generic APIAny agent (.md)

Trust score — 8/10

/10 trustVerified

Scored 8/10 (verified tier, metrics-v1). Evidence: 1,810 GitHub stars (log-scaled vs catalog peers); last push 0d ago; 100 releases/12mo; 29 open issues. Manual review: thin independent coverage; funding/cold-start figures unverified. License: AGPL-3.0.

★ 1,810 GitHub stars

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
6.4/10
Maintenance · recency, releases, issue load
9.4/10
Openness · license + docs
10.0/10
Manual review · sanity read + red-flag check
7.0/10
  • · 1,810 GitHub stars (log-scaled vs catalog peers)
  • · last push 0d ago; 100 releases/12mo; 29 open issues
  • · manual review: thin independent coverage; funding/cold-start figures unverified

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.