Agent sandboxes / orkestr Sandboxes alternative

An orkestr Sandboxes alternative for the state your agent retains

orkestr Sandboxes provides EU-operated VMs and a pause/resume workflow that can continue background processes. Its /persist volume is separate from the sandbox’s writable overlay and RAM-backed workspace. Choose the correct retention path before comparing long-running agent machines.

Start buildingCompare capabilities ↓

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

When orkestr Sandboxes fits

orkestr fits teams prioritizing its EU operator and regions, separate-kernel machines, per-sandbox networking, and actual-consumption billing. Its mounted volumes can retain files after a sandbox terminates, while pause/resume can preserve a live environment between requests.

When to consider Sandbox0

Consider Sandbox0 when one complete durable RootFS should expand online across tasks and supported file access should leave compute stopped. Evaluate jurisdiction, kernel compatibility, process requirements, and actual workload resource use independently.

Capabilities

orkestr Sandboxes and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeorkestr SandboxesCarded tiers have no continuous-running cap; per-call countdown and account limits still apply.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 filesorkestr SandboxesThe /persist volume survives termination; pause retains a different sandbox snapshot with plan-specific expiry.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSorkestr SandboxesPause can continue background commands, but the reviewed guides do not establish an unconditional durable complete-RootFS contract across all lifecycle paths.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 stateorkestr SandboxesPause snapshots the sandbox; resume consumes that snapshot and restarts the lifetime clock, subject to retention/admission.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateorkestr SandboxesThe product FAQ explicitly says snapshots include memory; the SDK documents continuing background processes.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.

orkestr’s documented persistent volume defaults to 10 GB and is capped at 100 GB per volume, with provisioned-capacity billing and one attached sandbox at a time. That mount and a paused snapshot are distinct from automatically expanding an entire durable RootFS while the agent runs.

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.

The orkestr SDK says auto_resume makes exec and file calls wake paused compute, including presigned file access. This differs from Sandbox0’s supported persistent Files API operations that keep guest compute stopped.

See paused file operations →

Pause retention and terminate retention have different scope

Background commands can survive pause/resume, but paused snapshots expire after seven days on Free or 30 days on carded accounts; Enterprise has no expiry per current limits. Termination drops the writable overlay and RAM-backed /workspace and /tmp. Files that must survive termination belong on /persist, whose volume has its own lifecycle.

Source 1 ↗Source 2 ↗Source 3 ↗Source 4 ↗

Continuous runtime, countdown and command timeout are separate

The limits guide allows uncapped continuous running on carded tiers, while Free has a one-hour running-window cap. A single countdown is still bounded, and each command has a separate timeout. The FAQ says lifetime extension is unavailable, but the current SDK and limits guide document set_timeout. Check the effective API rather than treat that stale FAQ as an absolute prohibition.

Source 1 ↗Source 2 ↗Source 3 ↗

Sandbox EUR rates are separate from the application-platform subscription

The public Sandboxes price tab meters actual on-CPU time and working-set RAM rather than reserved allocation. Its storage meter is separate. The monthly per-seat Platform subscription is a different product and is not required for pay-as-you-go Sandboxes. Compare the actual consumption and storage basis without converting the directory’s historical USD estimates into current EUR prices.

Source 1 ↗Source 2 ↗

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.

orkestr Sandboxes

Pricing reviewed 2026-10-18

Consumed CPU
€0.045 per vCPU-hour
Actual on-CPU time, metered per second.
Rate source ↗
Working-set RAM
€0.015 per GiB-hour
Rate source ↗
Retained storage
€0.10 per GB-month
Snapshots, templates and volumes share storage billing; volume capacity is provisioned, not only used bytes.
Rate source ↗
Sandbox platform fee
No monthly sandbox platform fee
Rate source ↗
Published storage allowance
First 10 GB of held snapshot/template storage free
Confirm allowance coverage for the chosen storage type.
Rate source ↗

Keep EUR, GiB RAM and GB storage labels; no exchange-rate conversion or equal-throughput claim is made. Marketing concurrency/session tables, the FAQ and current limits differ; inspect account-effective limits. Pausing stops compute, but retained storage continues, and a Platform seat subscription is a separate product.

Check current orkestr Sandboxes 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 orkestr Sandboxes

  1. Inventory writable-overlay changes, /workspace versus /persist data, paused expiry, effective account limits, egress settings, and actual consumption/storage billing.
  2. Export required files and port compatible templates, commands, lifecycle, credentials, network rules, and application services.
  3. Validate system-path retention and growth, paused file access, required process behavior, jurisdiction and capacity, then calculate all effective compute and storage charges.

Common questions

Is Sandbox0 a drop-in orkestr Sandboxes 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 ↗orkestr.eu/docs/sandboxes/python-sdk.md ↗orkestr.eu/docs/sandboxes/limits ↗orkestr.eu/docs/sandboxes/faq.md ↗orkestr.eu/docs/sandboxes/volumes.md ↗orkestr.eu/pricing ↗orkestr.eu/sandboxes ↗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 →