CompareSandboxes
Get the brief
AI code-execution sandbox

Northflank

Northflank runs sandboxes with selectable Kata/gVisor microVM isolation, unlimited session duration, any OCI image and BYOC deployment — at the lowest published vCPU rate in the category.

Last verified 2026-07-26

Visit Northflank ↗ Get pricing & demo

Northflank positions as the best-overall and cheapest option ($0.01667/vCPU-hr), with a choice of Kata or gVisor isolation, unlimited session length, any OCI image, BYOC/VPC across major clouds, and GPUs. It processes 2M+ isolated workloads a month. (Note: Northflank also publishes many of the category comparison posts.)

Key specs

IsolationKata / gVisor microVM (selectable)
Cold start300ms
Price / vCPU-hr$0.01667
Pricing modelUsage-based
GPUH100 ~$2.74/hr
Session limitUnlimited
PersistencePersistent volumes
Languagesany (OCI image)
MCP support—
IntegrationsREST/API
DeploymentManaged + BYOC/VPC (AWS/GCP/Azure/Oracle)
Free tierFree trial
vCPU rate$0.01667/hr
Throughput2M+ workloads/mo

Strengths

  • Lowest published vCPU rate
  • BYOC/VPC
  • Unlimited sessions
  • Any OCI image + GPU

Trade-offs

  • Less agent-SDK-specific tooling
  • Broader platform, steeper setup

Best for: Cost-sensitive teams and enterprises needing BYOC and unlimited sessions.

Head-to-head

Northflank vs E2B →Northflank vs Modal →

Alternatives to Northflank

E2B — AI code-execution sandboxDaytona — AI code-execution sandboxModal — AI code-execution sandboxVercel Sandbox — AI code-execution sandbox

Sources: Northflank pricing ↗

Get pricing for Northflank

Tell us what you're building. We'll route you to the right sandbox teams for pricing and a demo — no spam.