AI Potluck
Infrastructure / Deployment

Beta9

Beam Cloud

Open-source Docker-based sandbox runtime positioned as a self-hostable E2B alternative, with first-class GPU support, Python and Node.js runtimes, and no session time limits. Picked when teams need to keep agent execution on their own infra (HIPAA/SOC2/BYOC) and want GPU workloads alongside CPU sandboxes. ~1.6k stars, AGPL-3.0 licensed with a managed cloud offering.

Beta9, open source serverless AI runtime (sandboxes, serverless inference endpoints, task queues); worker v0.1.597 (Jun 4 2026) confirmed live on GitHub. OSS engine powering managed Beam cloud (beam.cloud).

Openness

4 high confidence
4.0
license
AGPL-3.0(OSI, copyleft)
source
public
managed-tier
Beam (beam.cloud) fully-managed GPU cloud (proprietary)
core-gated
gated

AGPL-3.0 (OSI copyleft) engine plus the proprietary managed Beam cloud => open_core. Scored 4 on the same footing as the Apache-licensed open-core peers. Corrected from 3, which had followed Daytona in penalizing copyleft; both are corrected together.

Adoption

2 low confidence
2.0

No PyPI/usage figure for the OSS runtime surfaced and no disclosed Beam customer/usage count; only ~1.7k GitHub stars available as a signal. stars_fallback caps level at 3; placed at 2 given the small star base and absence of any volume signal. Low confidence.

Capability

3 medium confidence
3.0

Capable serverless container runtime with GPU support, volumes, autoscaling and sandbox primitives, but container-level (not VM/microVM) isolation places it below the VM-grade peers on the isolation-strength dimension; mid-tier. Independent ComputeSDK TTI benchmarks of Beam (the managed beta9 cloud) corroborate the fast container launch (~380ms median cold-start, 8th of 19 sandbox providers).

  • https://github.com/beam-cloud/beta9 recorded 2026-06-04

    serverless runtime: rapid container launches, parallelize across hundreds of containers, GPU support, volume storage, sandboxes

  • https://www.computesdk.com/benchmarks/sandboxes/ recorded 2026-07-09

    ComputeSDK TTI leaderboard (independent, 100 iters/day, 2026-07-09 run): Beam (managed beta9) median time-to-interactive ~380ms sequential / ~810ms burst @100 concurrent; 8th of 19 providers

Unchanged since 2026-07-30 (last edited, not re-checked)