Computer-use platforms / H Company / Surfer / H Agents API alternative

H Agents API alternative

H Company’s Agents API runs computer-use agents with a model and harness, while its Models API supplies inference separately. Browser profiles and local-device control add further choices. Compare those managed agent capabilities with the runtime that owns your application’s Linux tools and files.

Start buildingCompare capabilities ↓

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

When H Company / Surfer / H Agents API fits

H fits teams that want a managed computer-use agent, cloud browser or control of an existing desktop, with built-in session monitoring and model integration. Its Models API and on-device products have different operating responsibilities from managed Agents API sessions.

When to consider Sandbox0

Consider Sandbox0 when you need to operate your own compatible agent stack and retain system-package changes, configuration and growing RootFS across tasks. You must supply or retain the model, agent loop and computer-use tools; raw compute prices are not a quote for H’s complete agent service.

Capabilities

H Company / Surfer / H Agents API and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeH Company / Surfer / H Agents APINot 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 filesH Company / Surfer / H Agents APIBrowser profiles persist selected web state; a managed general-purpose durable Linux disk is not established.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSH Company / Surfer / H Agents APICloud browser/local-device control does not establish a whole writable Linux RootFS service.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 stateH Company / Surfer / H Agents APIBrowser profile writeback is not arbitrary-machine pause and memory restore.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateH Company / Surfer / H Agents APIArbitrary process-memory capture/restore is not established by these environment interfaces.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.

H browser profiles save selected web state. The reviewed environment/profile contract does not establish automatic expansion of an entire Linux RootFS for a customer’s arbitrary toolchain. Sandbox0 provides that execution/storage layer around an agent implementation.

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.

Managing a stored H browser profile is not a Files API for reading and modifying the whole paused Linux RootFS. A locally controlled desktop inherits its host’s storage policy rather than acquiring Sandbox0-style paused-file behavior through the agent API.

See paused file operations →

Local control and running the agent locally are different modes

The environment guide distinguishes H-hosted cloud browsers from a browser or desktop on the user’s device. Local control keeps the agent in H’s cloud; it moves only the controlled interface to your machine. Running the agent itself locally uses a separate local mode. Make that boundary explicit when moving workloads or evaluating data flow.

Source 1 ↗

Default-profile concurrency has a write policy

Profiles retain browser cookies, local storage and session data until deletion. Default-profile sessions save final state on completion, best-effort. When several run concurrently, the first to start writes back and the others run read-only. Reusable browser state is useful, but it is not a full Linux filesystem or RAM checkpoint.

Source 1 ↗

Session counts are estimates of model consumption

The public Agents API cards combine a fee, concurrency and model spend. The listed approximately 60, 1,000 or 10,000 sessions vary with task complexity and model selection. Builder and Startup top-ups add 20% over model cost after the initial model-cost allowance; do not read the session examples as guaranteed flat-price executions.

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.

H Company / Surfer / H Agents API

Pricing reviewed 2026-10-18

Agents API Free
$0/month
Up to three concurrent agents; approximately 60 sessions, capped by token fair usage.
Rate source ↗
Agents API Builder
$20/month
Up to ten concurrent agents; initial $20 at model cost, then model cost + 20% for top-ups.
Rate source ↗
Agents API Startup
$200/month
Up to 100 concurrent agents; initial $200 at model cost, then model cost + 20% for top-ups.
Rate source ↗
Enterprise
Custom rates and deployment terms
Negotiated model/markup, concurrency and infrastructure terms.
Rate source ↗

The quoted session counts on the public cards are workload-dependent estimates. These rows describe managed Agents API plans, not Models API per-token rates, HoloTab extension pricing or the cost of operating local hardware. Concurrency, model fair usage, proxies and deployment terms affect the comparison.

Check current H Company / Surfer / H Agents API 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 H Company / Surfer / H Agents API

  1. Identify managed Agents API versus model-only or local-control usage; inventory profiles, writeback concurrency, skills and required session outputs.
  2. Implement or retain the agent/model loop while moving compatible Linux tools and files to Sandbox0, validating headed-browser or desktop requirements separately.
  3. Test concurrent profile behavior, task cost at actual model usage, RootFS retention/growth, paused files and full model plus runtime cost.

Common questions

Is Sandbox0 a drop-in H Company / Surfer / H Agents API 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 ↗hub.hcompany.ai/agents-api/environments/overview ↗hub.hcompany.ai/agents-api/browser/profiles ↗hcompany.ai/pricing ↗Sandbox0 RootFS capacity →

Related comparisons

Bytebot vs Sandbox0 →CloudAxis vs Sandbox0 →Coasty vs Sandbox0 →ComputerUse.space vs Sandbox0 →Cyberdesk Windows Automation vs Sandbox0 →Google Agent Platform Sandbox (Code Execution / Shell / Computer Use) vs Sandbox0 →