Development environments / OpenHands Remote Sandbox / Cloud alternative

An OpenHands remote sandbox alternative at the execution-provider boundary

OpenHands remote sandbox mode runs the agent server on another execution provider. Its Workspace abstraction spans local processes, Docker and remote servers, while OpenHands Cloud is a specific managed implementation. Compare that agent/execution integration with the durable machine service behind it.

Start buildingCompare capabilities ↓

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

When OpenHands Remote Sandbox / Cloud fits

Keep OpenHands when its agent orchestration, tools, conversation model and server protocol are useful to your application. A change in underlying execution infrastructure may be an integration project rather than a reason to replace the agent framework itself.

When to consider Sandbox0

Consider Sandbox0 for compatible Linux execution with a persistent complete writable RootFS, automatic online growth and supported paused file operations. Validate an adapter to the required OpenHands server/workspace contract rather than assume SDKs are interchangeable.

Capabilities

OpenHands Remote Sandbox / Cloud and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeOpenHands Remote Sandbox / 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 filesOpenHands Remote Sandbox / CloudRemote-provider configuration does not itself establish durable file retention; inspect the concrete implementation.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSOpenHands Remote Sandbox / CloudCommand/file APIs do not establish complete RootFS retention for every Workspace implementation.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 stateOpenHands Remote Sandbox / CloudCloud keep_alive retains a running sandbox; a compute-release pause contract is not established by these guides.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateOpenHands Remote Sandbox / CloudThe reviewed Workspace abstraction does not promise arbitrary process-memory capture/restore.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 Workspace protocol describes execution and file transfer, not automatic online expansion of a complete durable RootFS. Validate storage continuity and growth on the actual backend selected for the agent server.

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 →

Remote mode is a provider integration, not one resource contract

The remote sandbox guide requires a remote API and credentials, whose configuration varies by deployment. The architecture guide abstracts commands and file operations across multiple implementations. A generic remote workspace therefore does not establish one disk-retention, kernel, pause or memory-checkpoint guarantee.

Source 1 ↗Source 2 ↗

The Cloud workspace deletes its sandbox on cleanup by default

OpenHandsCloudWorkspace provisions and monitors a managed sandbox. Its keep_alive option defaults to false, so closing the workspace normally deletes the sandbox; enabling it keeps the sandbox running for later inspection. Keeping it running is not the same as releasing compute while retaining a durable machine.

Source 1 ↗

Price the actual backend and agent stack

OpenHands lists a free Cloud Individual plan, with separate model usage, and custom Enterprise pricing. Its remote integration can also target other deployments with their own costs. Compare the chosen product plan, underlying execution provider, model charges and retained storage rather than assume one remote-sandbox price applies to every Workspace implementation.

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

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.

OpenHands Remote Sandbox / Cloud

Pricing reviewed 2026-10-18

Generic remote sandbox rate
Deployment/provider-specific; no universal numeric rate in this guide
Rate source ↗
OpenHands Cloud Individual
Free plan; model usage separate
One user and 10 daily conversations in the current plan comparison; BYOK or OpenHands model providers at cost.
Rate source ↗
OpenHands Enterprise
Custom pricing
Rate source ↗
Custom remote deployment
Underlying compute, storage, model and operations costs apply
Rate source ↗

The public Cloud Individual plan is free with separate model usage; Enterprise is custom-priced. These product plans do not establish a universal remote-provider compute rate. A self-operated backend still has infrastructure and operations costs.

Check current OpenHands Remote Sandbox / 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 OpenHands Remote Sandbox / Cloud

  1. Identify the concrete Workspace/provider, cleanup behavior, server protocol, models, credentials, files and conversation/artifact storage.
  2. Build and test the necessary execution-provider adapter and compatible agent-server image on Sandbox0, while retaining required OpenHands behavior.
  3. Validate command/file semantics, whole-system retention, growing outputs, paused file access, process expectations and the combined runtime/model/product bill.

Common questions

Is Sandbox0 a drop-in OpenHands Remote Sandbox / Cloud replacement?

Evaluate API compatibility and product scope. These products serve different core workflows, so a change may require a different architecture.

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.openhands.dev/openhands/usage/sandboxes/remote ↗docs.openhands.dev/sdk/arch/workspace.md ↗docs.openhands.dev/sdk/guides/agent-server/cloud-workspace.md ↗www.openhands.dev/pricing ↗Sandbox0 RootFS capacity →

Related comparisons

Agency Tool Company vs Sandbox0 →Amp Orbs vs Sandbox0 →Bitrise RDE vs Sandbox0 →boxes.dev vs Sandbox0 →Capy vs Sandbox0 →Claw 2 Agent vs Sandbox0 →