Workflow orchestration / Hyrex alternative

Hyrex task queue alternative

Hyrex is a PostgreSQL-backed task framework with queues, retries, timeouts and workflow dependencies. It retains task coordination state while workers execute functions. The environment hosting those workers determines whether installed tools and local filesystem changes survive.

Start buildingCompare capabilities ↓

Reviewed 2026-10-19 · Published by the Sandbox0 team

When Hyrex fits

Choose Hyrex for Python or TypeScript task orchestration, queue routing and recurring jobs with a Postgres-backed control plane. Retain those workflow semantics when agents need retries, dependencies and scheduling.

When to consider Sandbox0

Consider Sandbox0 for the compatible worker’s durable Linux environment. Task scheduling, database durability, idempotency, retries and monitoring remain orchestration responsibilities you retain or implement.

Capabilities

Hyrex and Sandbox0: what persists?

Feature comparison with documentation and verification status
Environment and runtime lifetimeHyrexNot 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 filesHyrexPostgres and task/KV records are documented; worker disk durability is separately operated.View evidence ↗Sandbox0The writable RootFS survives successful checkpoints, including software, system configuration, dependencies, and user data.Read documentation →
Whole RootFSHyrexTask orchestration does not establish a managed whole writable RootFS lifecycle.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 stateHyrexTask retries and scheduling are distinct from machine checkpoint/resume.View evidence ↗Sandbox0Pause compute between tasks; resume on available capacity. Default resume starts fresh processes from the saved filesystem.Read documentation →
Memory and process stateHyrexNo arbitrary process-memory checkpoint is established.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.

Hyrex retains application coordination in Postgres. RootFS capacity and system-tool retention belong to each worker backend; Sandbox0 can provide a durable growing Linux environment underneath the task framework.

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.

Database task records can remain available without a live worker. That is a different interface from reading and mutating supported files throughout a paused worker RootFS.

See paused file operations →

Durable tasks are application state

The Python SDK documents task metadata, retries, dependencies, cron scheduling and a distributed key-value store. These are useful coordination primitives, but they do not checkpoint an arbitrary worker’s installed system packages, complete RootFS or process memory.

Source 1 ↗

Worker hosts remain an explicit part of deployment

The SDK initializes a Postgres or Cloud connection and starts workers as separate processes. Hosting a worker in a container does not automatically give it durable machine state; retain files through the execution backend and task results through the chosen database.

Source 1 ↗

Separate software licensing from hosted service terms

The Python SDK is Apache-2.0 licensed. Running it yourself still requires Postgres, worker compute, storage and operations. Confirm current Hyrex Cloud availability, allowances and commercial terms with the vendor; no numeric hosted task or resource tariff is established here.

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.

Hyrex

Pricing reviewed 2026-10-19

Python SDK
Apache-2.0 software
Self-operated database, workers, storage and operations remain separate.
Rate source ↗
Managed service
Confirm current vendor terms
No numeric hosted rate or universal free worker allowance is established.
Rate source ↗

A software license does not allocate free managed compute. Verify current Cloud access and pricing before integrating; no service shutdown or ongoing numeric offer is inferred.

Check current Hyrex 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 Hyrex

  1. Inventory task definitions, queues, cron jobs, retry/idempotency policies, Postgres and worker ownership.
  2. Retain the orchestration layer and place compatible workers in Sandbox0 with explicit credentials and lifecycle integration.
  3. Test recovery of tasks alongside system-tool retention, growing files and total database, worker, storage and model costs.

Common questions

Is Sandbox0 a drop-in Hyrex 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 ↗github.com/hyrex-labs/hyrex-python ↗hyrex.io/cloud ↗Sandbox0 RootFS capacity →

Related comparisons

Expanse Cluster Intelligence vs Sandbox0 →Google Agent Runtime vs Sandbox0 →Hatchet vs Sandbox0 →Mastra vs Sandbox0 →Microsoft Foundry hosted agents vs Sandbox0 →Nextmv DecisionOps vs Sandbox0 →