Kestra hosting

Declarative orchestration for data and AI pipelines.

AutomationApache-2.0 (community edition)from $11.99/mo

In short

Kestra hosting at AIHostBox means a dedicated, single-tenant Kestra instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $11.99 per month and the admin account, the data and any API keys are yours.

YAML workflowsGit-friendlyPluginsBackfillsParallel tasks

Kestra is an event-driven orchestration platform from Kestra Technologies, Apache-2.0 in its open-source edition. Every workflow is a YAML flow of tasks with inputs, triggers, retries and error handling, which you edit in the built-in code editor, keep in Git and review like any other code. Tasks come from hundreds of plugins covering databases, cloud storage, messaging and scripts in Python, Node.js, R, Go, Shell and SQL. That declarative, code-first model sets it apart from drag-and-drop tools such as Activepieces or Node-RED.

AI now sits inside the same model: AI agent and RAG tasks call OpenAI, Anthropic, Gemini, Bedrock or a local Ollama as steps in an ordinary flow. An orchestrator has to be awake when its schedules fire and webhooks arrive, and it keeps a full execution history — logs, outputs and exactly which task failed — that the whole team needs to see. That is server work: Kestra runs as a Java service, with Postgres holding flows, the execution queue and state, and internal storage holding the files tasks produce.

Kestra hosting plans

PlusRecommended minimum

4 GB RAM · 2 vCPU · NVMe

$11.99/mo

Order

ProTeams & heavier load

8 GB RAM · 3 vCPU · NVMe

$18.99/mo

Order

Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.

How we run Kestra

The kestra/kestra server in standalone mode on port 8080, with Postgres as both repository and queue and local internal storage on a persistent volume, as in the project’s docker-compose.

What we configure for you

kestra.repository.type and kestra.queue.type set to postgres, the production backend upstream recommends
kestra.storage.type local on persistent disk, or s3 if you want task outputs in your own bucket
Basic-auth credentials for UI and API, mandatory since 0.24
kestra.encryption.secret-key so SECRET-type inputs and outputs work
kestra.url set to your HTTPS hostname
SECRET_-prefixed environment variables for your API keys

Which plan fits

Plus (4 GB, 2 vCPU) suits a team running scheduled pipelines with modest parallelism; the JVM and Postgres take a fair share of it at rest. Pro (8 GB, 3 vCPU) is for many concurrent executions, heavy script tasks or long execution histories. We do not sell a 2 GB plan for Kestra because the Java server alone makes it cramped.

The honest limits

SSO, RBAC, multi-tenancy, external secret managers and Kafka-backed scaling are Enterprise Edition only. Every task runs on the plan’s CPU, so heavy data processing is bounded by its vCPUs.

What people run Kestra for

ETL and ingestion

Scheduled extraction into your warehouse or vector store.

Reproducible pipelines

Workflow definitions that live in your repository.

Scheduled AI jobs

Run an agent or summarisation task on a schedule and store the output.

Event-driven automation

Start a flow from a webhook, a file landing in storage or a message on a queue.

Licence and version policy

Item	Position
Upstream licence	Apache-2.0 (community edition)
Sold as managed hosting	Yes
Tenancy	Single tenant — one customer per instance
Patching	Critical advisories within 24 hours, high severity within 72

Our full position on licences, including the applications we decline to sell, is on the licensing page.

Reading on Kestra

Comparisonn8n alternatives you can host commercially Guide · AutomationHow to self-host n8n (and what the licence actually says)

Works well with Kestra

More Automation we host

Questions

Kestra hosting: your questions

How do people log in to open-source Kestra?

Since 0.24, basic authentication is mandatory in the open-source edition: on first visit you create a username and password on the setup page, or we set them in configuration. SSO, RBAC and multi-tenancy are Enterprise Edition features, so the open-source instance suits a single team sharing access.

Can I keep my flows in Git?

Yes, and it is the recommended way to work. Flows are YAML files, and Kestra’s Git plugin tasks can sync flows and namespace files from a repository, or push changes made in the UI back to one. You can also deploy flows from CI through Kestra’s API.

Where are secrets kept in the open-source edition?

As environment variables prefixed with SECRET_ and base64-encoded, read in a flow with the secret() function, so API keys never appear in flow YAML. We add them to your instance on request. External secret managers such as Vault are an Enterprise Edition feature.

What does a Kestra backup contain?

Flow definitions, execution history, logs and the queue live in Postgres; files produced by tasks live in internal storage on the instance. Nightly off-box backups cover both, so a restore brings back flows and past executions. Keeping flows in Git as well gives you a second, independent copy of the source.

Can a Kestra flow call my own local model?

Yes. The AI plugins support Ollama alongside OpenAI, Anthropic, Gemini and Bedrock, so a flow can send text to a model on your own Ollama plan over HTTPS and pass the result to the next task. Keep the endpoint key in a secret rather than in the flow.

Run Kestra without running a server

A dedicated Kestra instance, deployed and maintained by us. Month to month, cancel whenever.

Browse the catalogue Talk to a human