Agent sandboxes / Unikraft Cloud alternative

A Unikraft Cloud alternative for a growing Linux agent computer

Unikraft Cloud supports stateful microVMs and a Sandbox SDK, alongside its application platform. Its root image format, writable volumes, snapshots, and automatic scale-to-zero determine what an agent keeps. Check that storage layout before treating an instance as a mutable general-purpose computer.

Reviewed 2026-10-18 · Published by the Sandbox0 team

When Unikraft Cloud fits

Unikraft Cloud fits teams using its microVM application stack, stateful scale-to-zero, instance branching, plugins, and network shield. The sandbox API covers command execution and files, and the platform supports both specialized runtimes and Linux-based microVMs.

When to consider Sandbox0

Consider Sandbox0 when installed tools, configuration, and accumulating outputs should live in one writable durable RootFS with automatic online expansion. Its runtime offers a Linux agent environment without requiring the application to design separate persistence for each system path.

Capabilities

Unikraft Cloud and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeUnikraft CloudNot established in this reviewSandbox0ttl and hard_ttl default to 0 (disabled). Configure idle pause and deletion policies for the same persistent sandbox identity; an idle running sandbox still accrues compute charges.Read documentation →
Persistent filesUnikraft CloudPersistent volumes survive restart and redeployment; deletion behavior depends on how they were created.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSUnikraft CloudRoot image format, writable volumes, and retained instance state must be considered together; no single whole-writable-RootFS guarantee is established here.View evidence ↗Sandbox0Persistence extends beyond /workspace to paths such as /usr/local, /etc, and /home. Runtime tmpfs and virtual filesystems are excluded.Read documentation →
Release compute, keep stateUnikraft CloudStateful scale-to-zero snapshots and resumes instance execution.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateUnikraft CloudStateful mode retains memory for resumed execution; stateless cold boot is a distinct choice.View evidence ↗Sandbox0Experimental memory checkpoints can preserve compatible process state. Use memory: true on both pause and resume on a compatible deployment. A filesystem checkpoint alone does not preserve running processes.Read documentation →

Online RootFS growth

On growth-enabled images, Sandbox0 automatically expands the writable RootFS as free space runs low, within a configured sparse logical ceiling up to 64 TiB. Agent processes keep running. Growth is enabled in the qualified Cloud region; confirm activation for other regions. Existing images and snapshots retain their captured storage geometry. The ceiling is logical capacity, not preallocated host disk.

The reviewed rootfs and volume guides expose image-format and mounted-storage choices. Stateful snapshots and platform vertical scaling do not by themselves establish automatic free-space-triggered growth of an entire durable writable RootFS.

Files while compute is paused

Sandbox0's Files API reads persistent files without resuming compute. Filesystem-only pauses also permit edits; retained-memory pauses are read-only and reject mutations with 409. SSH, SFTP, file watching, and runtime-only mounts require running compute.

See paused file operations →

Stateful standby and deletion have different consequences

Unikraft stateful scale-to-zero saves an instance snapshot and resumes execution on incoming traffic. The Sandbox SDK’s await-using example instead deletes the sandbox at scope exit. Decide whether the application is retaining a reusable computer or intentionally destroying a task environment.

Choose image format and writable storage deliberately

The rootfs guide distinguishes memory-unpacked CPIO from read-only EROFS. Persistent volumes keep data across restarts and redeployments, but volumes created with an instance can be tied to its deletion. Those choices matter for installed packages and files outside the selected mount paths.

Scale-to-zero and larger resource quotas solve different problems

A network-idle instance can sleep even while background work needs to continue, so Unikraft documents temporarily disabling scale-to-zero. Paid plans also increase running, standby, memory, image, and storage quotas. Those quotas are not a published per-GiB-hour price or proof that every root disk automatically expands.

Pricing

Compare billing models, then price your workload

CPU allocation, memory, billing minimums, disk, plan fees, and supported workload differ. These meters are not equal-performance machine quotes. Confirm region, resource units, and included allowances before estimating a total.

Unikraft Cloud

Pricing reviewed 2026-10-18

Hobby plan
$0 per month
Lists 2 running and 5 standby instances, with resource quotas.
Rate source ↗
Team plan
$39 per month
Lists 8 running and 50 standby instances.
Rate source ↗
Pro plan
$199 per month
Lists 24 running and 5,000 standby instances.
Rate source ↗
Enterprise
Custom terms
Rate source ↗

These are the current public plan fees, not invented per-resource consumption rates. Compare CPU, memory, standby storage, image storage, and feature entitlements for the actual fleet. The pricing page advertises zero idle cost; retained-state and storage quotas still constrain the plan.

Check current Unikraft Cloud pricing ↗

Sandbox0

$0.015/GiB-hour compute (billed by the second at approximately $0.000004167/GiB-second), plus $0.02/GiB-month of retained incremental RootFS and memory checkpoint data. Shared base images and platform recovery retention are excluded; paused sandboxes accrue storage charges only.

Compute / GiB-hour
$0.015
Retained data / GiB-month
$0.02
Paused compute
$0.00

A 4 GiB agent, 100 running hours

$6.00 compute

4 GiB × 100 hours, billed by the second. Keeping 20 GiB of billable incremental data for a full billing month adds $0.40. CPU is derived from memory by the platform; choose a tier that fits your workload.

Understand storage billing →

Before migrating from Unikraft Cloud

  1. Inventory metro, runtime kernel, rootfs format, volume ownership and deletion policy, snapshots, templates, and plugin configuration.
  2. Export mutable state from its actual volumes or retained instance, and port commands, files, service exposure, and network-shield credentials.
  3. Validate package compatibility, background tasks under the intended idle policy, whole-RootFS persistence and growth, and supported paused file operations.

Common questions

Is Sandbox0 a drop-in Unikraft Cloud replacement?

Evaluate API compatibility and product scope. A shared sandbox category does not make SDKs, lifecycle semantics, or limits interchangeable.

Does keeping files also keep a running agent?

Filesystem state and volatile process state are separate. Sandbox0 requires explicit compatible memory checkpoints to preserve running processes; its default filesystem resume starts fresh processes.

Sources and review scope

The Boat team's Battleships directory is the discovery source. References below describe the product and plan discussed on this page. Check current availability, regional terms and billing units before migrating. Where the cited material does not establish a capability or price, the comparison identifies that uncertainty.

Related comparisons