Announcing New Pay-As-You-Go Pricing →

Only pay for what you run.

We set out to build the most performant orchestration engine on the market. The result: lower prices as you scale.

Pay-as-you-go

$0

to start
Only pay for what you run with a generous free tier.
  • 1M runs and 1M events free, every month
  • Then $30 per million runs, $5 per million events
  • 100 workers, 10 tenants, 50 users
  • 7-day retention, 10 GB/day active storage
  • SOC 2 Type II, community support
Start building free
Custom

Let’s talk

For 100M+ runs a month, dedicated capacity, or enterprise features.
  • Volume discounts
  • Latency guarantees and SLAs
  • Dedicated capacity and multi-region
  • SSO, RBAC, audit logs, HIPAA
  • Self-hosted and BYOC support
Talk to an engineer
Want to self-host? Hatchet is 100% open-source and MIT licensed. Learn more

Cost estimate

Drag the sliders to estimate your cost.

10k100k1M free10M100M1B
10k100k1M free10M100M1B
Payload size
Payloads count as active storage for one day. The first 10 GB each day is free.
Logs and seatsOptional
50 included, then $10/user
10 GB free, then $0.0004/GB-hr
Pay as you go

$0

/ month
Task runs1M inside the free million
$0
Events200k inside the free million
$0
Active storage1M × 2 KB ≈ 64 MB a day, inside the free 10 GB
$0
Estimated total$0/mo
Start building free

Full comparison

Full breakdown of each plan.

Pay-as-you-goNo card
Pay-as-you-goCard on file
Custom
Features
Task runsOne execution of a task. A task can spawn another task, which counts as two task runs.
2,000/day
1M/mo includedthen $30/M
Volume discounts
EventsEvents pushed to Hatchet that trigger workflows or satisfy wait conditions.
2,000/day
1M/mo includedthen $5/M
Volume discounts
Scheduled runsOne-time tasks scheduled for a specific time in the future. The limit is how many are scheduled at any given moment in time.
100,000
1,000,000
Custom
WorkersProcesses you run that connect to Hatchet and execute tasks. The limit is how many workers can be connected at once.
10
100
Custom
Worker slotsEach slot runs one task at a time. A worker with 100 slots can run 100 tasks concurrently. The limit is the total across all workers.
2,000
10,000
Dedicated capacity
Queue backlogTasks waiting to run at any moment. Once the backlog is full, new tasks are rejected until it drains.
10,000 tasks
1,000,000 tasks
Custom
WebhooksHTTP endpoints that can trigger runs in Hatchet. The limit is how many unique endpoints you can configure. Each webhook produces an event.
10
100
Custom
Cron jobsRecurring schedules that start a workflow on a cron expression.
10
1,000
Custom
Storage and retention
Active storageTask inputs and outputs held for 24 hours. Storage above the included amount is billed per GB-hour.
500 MB included
10 GB includedthen $0.03/GB-hr
Custom
Log and trace ingestionApplication logs and OpenTelemetry spans sent to Hatchet, measured by the volume ingested each month.
500 MB included
10 GB includedthen $0.0004/GB-hr
Custom, with export
RetentionHow long run history, logs, and traces stay queryable in the dashboard and API.
1 day
7 days
90+ days
Workspace
TenantsIsolated environments with their own workers, workflows, and API keys, such as dev, staging, and prod.
1
10
Custom
Users
3
50 includedthen $10/user
Custom
Platform
SOC 2 Type II
Included
Included
Included
DeploymentWhether the Hatchet engine is deployed and managed on Hatchet Cloud, or split into a hybrid or self-hosted setup.
Cloud
Cloud
Cloud, self-hosted, or hybrid
Performance tierWhether your workloads run on shared infrastructure or on capacity reserved for you.
Shared
Shared
Dedicated
Support
Discord
Discord
Dedicated Slack
SSO / SAML
n/a
n/a
Included
RBAC
n/a
n/a
Included
Audit logs
n/a
n/a
Included
HIPAA BAA
n/a
n/a
Included

Questions

Still unsure? Talk to an engineer.

Why is Hatchet cheaper than other orchestrators?

Hatchet is fast. Faster code means less load on the database, fewer spikes, and fewer machines. That’s what makes Hatchet more reliable, and it’s also what makes it cheaper.

What happens if I hit a free tier limit?

You will get alerted via email or Slack (if Slack alerts are enabled) as you are approaching your limit. If you exceed your limit, tasks will get dropped and you will see errors when trying to create new tasks.

Do failed runs count?

Yes. A run is an execution. If a task runs and fails, that execution counts.

Do retries count?

No. Hatchet doesn’t bill for retries. A task that fails and is retried counts as one run, however many attempts it takes.

When am I billed and how?

You are billed at the end of your billing cycle every 30 or 31 days. You pay nothing until you pass the included million runs and million events, and usage past that is billed monthly at the rates above.

I’m migrating from another orchestrator. Will you help?

Yes. We have migration guides for Temporal and Celery. You can also talk to an engineer.

Do you offer startup discounts?

Yes. Please reach out directly for details about our startup program.

Where does my data live?

Data plane regions in AWS Oregon, N. Virginia, and Ireland are live, and Sydney is in private beta. Custom plans can add dedicated capacity, more regions, or a self-hosted or bring-your-own-cloud deployment. See regions.

Start building free. Your first million runs and events each month are on us. No card needed to get started.

Run your first taskTalk to an engineer

  • No surprise bills. No plan fees. You pay $30 per million runs, only after your free million, with volume pricing at 100M+ runs a month.

  • Built for production. SOC 2 Type II on every plan, running on 20+ shards across 4 regions.

  • No lock-in. Open source and MIT licensed, you can self-host anytime.