Agent sandboxes / Superserve alternative

A Superserve alternative for an automatically growing agent RootFS

Superserve retains the whole VM state during pause and can create snapshots with memory and disk. Paused compute is not charged, and the default lifecycle has no automatic deletion deadline. The useful comparison is therefore storage growth, file access during pauses and billing conditions, rather than an assumed absence of persistent machines.

Start buildingCompare capabilities ↓

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

When Superserve fits

Superserve fits teams seeking full-state VM pause/resume, snapshots and preview endpoints, with configurable session deadlines and a paid enterprise private-cloud offer. Its generous retention defaults can suit agents that return to an existing execution state.

When to consider Sandbox0

Consider Sandbox0 when installed tools, settings and outputs should accumulate on an automatically growing RootFS and supported file operations should leave compute paused. Explicit compatible experimental memory pause is a separate choice from ordinary filesystem-only pause.

Capabilities

Superserve and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeSuperserveNo active-session timeout by default; optional timeoutSeconds applies to the entire active window.View evidence ↗Sandbox0ttl 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 filesSuperservePaused machines retain their filesystem with no default expiry; an optional cleanup policy can delete them.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSSuperserveSnapshots and pause retain VM disk plus execution state.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 stateSuperservePause checkpoints full VM state and stops compute billing; resume continues processes.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateSuperserveSnapshots capture memory and disk; captured process secrets require special handling.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.

Superserve’s public plans bound disk sizes, with different Free and Pro machine ceilings. Full-state snapshots preserve a configured machine; the reviewed lifecycle and pricing sources do not establish automatic online expansion of the complete RootFS. Evaluate separate storage integrations on their own terms.

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.

Superserve explicitly auto-resumes a paused sandbox for directory listing and other runtime file calls, and the list-directory documentation says it remains active afterward. Sandbox0’s supported paused Files API leaves guest compute stopped; memory-paused mutations return a conflict.

See paused file operations →

Pause preserves processes; timeout is an active-session cap

Superserve pause captures memory, processes and filesystem, then releases compute; resume continues that state. timeoutSeconds is total time in an active session, not an idle timer, so ongoing work does not reset it. Each resume starts a new timeout window. An unset timeout is disabled.

Source 1 ↗

Retention defaults differ from an optional paused-delete policy

Paused state is retained indefinitely by default. Optional autoDeleteSeconds counts one continuous paused interval, with a maximum setting of 30 days; zero deletes immediately. Resume cancels that countdown, and a later pause starts another. Do not confuse this optional cleanup timer with a universal maximum sandbox life.

Source 1 ↗

Snapshots retain execution state, but external credentials need care

Snapshots contain memory and disk and outlive the source until deleted. The guide warns that existing secret values inside captured processes belong to the original environment; secret-using processes may need restarting after restoration. A full-state image is not a replacement for reviewing credentials and external service relationships.

Source 1 ↗

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.

Same size. Lower compute cost.

2 vCPU · 4 GiB RAM · 730 running hours / month

Save $77.09 / month

The same CPU and memory allocation, with runtime aggregated across sessions. Monthly allowances and credits below are applied to this workload.

Superserve

Free · usage-based compute

$120.89

$0.1656/hour before monthly allowances

Sandbox0

Pay as you go · no monthly plan fee

$43.80

$0.06/hour · 63% lower in this example

Uses the public Pay-as-you-go resource rates for a running sandbox. Initial onboarding credits are excluded. Tracked storage is currently marked non-billable and is not added to this estimate.

Compute and stated plan fees only. Storage, network, taxes and one-time credits are excluded; hardware and throughput can differ.

Calculation and billing assumptions

Superserve: $120.888 compute after 0 included vCPU-hours and 0 included GiB-hours, plus $0.00 plan fee, minus $0.00 applied monthly credit.

Sandbox0: 4 GiB × 730 hours × its published compute rate. CPU uses the Cloud allocation shown on the pricing page.

Includes compute and the selected plan. Excludes storage, network, taxes, one-time grants and negotiated discounts. Assumes this workload can use the full listed monthly allowances. CPU and memory counts do not guarantee equal hardware or throughput.

Checked Superserve rate source ↗Sandbox0 rates ↗

Superserve

Pricing reviewed 2026-10-18

Active CPU
$0.0504 per vCPU-hour
Converted from the published $0.000014 per vCPU-second.
Rate source ↗
Active allocated memory
$0.0162 per GiB-hour
Converted from $0.0000045 per GiB-second.
Rate source ↗
Free plan
$0 plan fee; usage-based compute
$5 initial credit plus $95 after adding billing; these are onboarding credits.
Rate source ↗
Pro, annual-billing display
$12,000 billed annually
Displayed equivalent is $1,000/month including $1,000 monthly usage; this does not establish a monthly-payment offer.
Rate source ↗
Storage meter
Tracked, currently marked non-billable
The public rate object has billable:false; its tracked number must not be added as an actual charge.
Rate source ↗

The public rate card is effective from June 17, 2026. Active compute uses per-second resource rates; paused compute is zero. Current non-billable storage is not a permanent free-storage guarantee. Free and Pro resource ceilings differ, and enterprise private-cloud terms require a quote.

Check current Superserve 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
Understand storage billing →

Before migrating from Superserve

  1. Record active-session timeouts, paused cleanup timers, snapshots, machine sizing and all secret-using processes; export durable data before deleting the source.
  2. Recreate compatible tools, credentials, network and preview behavior, and decide whether each workload needs fresh processes or an explicit compatible memory checkpoint.
  3. Test repeated pauses, file-only access without wake-up, growing system packages and outputs, snapshot restoration and annual-plan versus measured compute cost.

Common questions

Is Sandbox0 a drop-in Superserve 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.

Battleships research record ↗docs.superserve.ai/sandbox/lifecycle.md ↗docs.superserve.ai/sandbox/snapshots.md ↗api.superserve.ai/billing/pricing/public ↗www.superserve.ai/pricing/ ↗Sandbox0 RootFS capacity →

Related comparisons

Agent 37 vs Sandbox0 →AgentComputer vs Sandbox0 →ainclave vs Sandbox0 →Alibaba Cloud Agent Sandbox / FC / AgentRun vs Sandbox0 →Alibaba Cloud AgentBay vs Sandbox0 →Amika vs Sandbox0 →