A Dagster alternative where self-hosted means the whole platform.
Dagster charges a credit for every asset materialized and every op executed, so a single pipeline run can rack up many credits depending on how many steps it touches. Serverless compute is billed separately too, by the minute, on top of those credits. Even their hybrid deployment option still runs through Dagster's own hosted control plane, not fully your infrastructure.
Unmeshed counts one flat number per workflow execution, and self-hosts completely, control plane included.
Trusted by teams at leading organisations
Overview
Two different approaches to orchestration.
About Unmeshed
Unmeshed is a workflow orchestration platform built by the team behind Netflix Conductor. You define workflows in JSON or YAML, with code-based workers in Java, Go, TypeScript, or Python. Retries, waits, and durable state come built in, and a single workflow execution costs the same flat number no matter how many steps run inside it.
About Dagster
Dagster is an asset-centric data orchestration platform with an open-source core, alongside the managed Dagster+ cloud offering. Pipelines are defined in Python as software-defined assets, and Dagster+ usage is billed in credits, one per asset materialized, and one per op executed, plus separate serverless compute charges by the minute. Dagster is currently in the process of joining Prefect.
Comparison
Unmeshed vs Dagster, feature by feature.
| Unmeshed | Dagster | |
|---|---|---|
| Platform | ||
| Built for | Code-first orchestration for developers and technical teams | Asset-centric data orchestration for data engineering teams |
| Workflow definitions | JSON, YAML, or code | Python only, defined as software-defined assets |
| Native integrations | 100+ | Native dbt, Snowflake, and Fivetran, plus a broad connector library |
| Visual designer | Included | |
| Resiliency and operations | ||
| Retry a single failed step without restarting the run | Included | Included |
| Hold, skip, rollback, restart as direct commands | Included | |
| Self-host the full control plane, not just compute | Included | Not included |
| Self-hosted deployment | Included | Included |
| Pricing | ||
| Pricing model | Flat, predictable plans | Credit per asset materialization and op, plus separate compute billing |
| Compute billed as a separate meter | Not included | Included |
| Free tier usable in production | Included | |
| Upgrade without a sales call | Included | Included |
| Security, compliance, and governance | ||
| Role-based access control | Included | |
| SSO / SAML | ||
| Audit logs | Included | |
| Self-hosted / on-premise deployment | Included | Included |
Count the run, not the steps inside it.
Start building workflows that cost the same flat number whether they run three steps or three hundred; on a free tier, you sign up for yourself.
Pricing
They count every step. We count the run.
Dagster charges one credit for every asset materialized and every op executed, so a pipeline with many steps costs more credits than a simple one. Serverless compute is billed separately by the minute, on top of that. Unmeshed counts one flat number: a workflow execution, no matter how many steps run inside it.
Free forever
$0/month
No credit card required
Get Started Free- 2,000 workflow executions / month
- Unlimited steps per workflow
- Community support
Premium
$20/month
Billed monthly. Annual billing available, saves 2 months
Get Started- 10,000 workflow executions / month
- Batch job scheduling
- Email support
Enterprise
Custom pricing
Contact Sales- Unlimited executions, workflows & seats
- SSO / SAML and role-based access control
- Dedicated support with SLAs
Dagster pricing, for reference
See Dagster's pricing pageSolo, $10/mo
Pay as you go, $0.040/credit, or $120/yr with 7,500 credits included
Starter, $100/mo
Pay as you go, $0.035/credit, or $1,200/yr with 30,000 credits included
Pro & Enterprise
Both custom pricing, contact sales. Serverless compute runs $0.010/minute on all tiers
Pricing details change often, treat this as a quick reference and confirm current numbers on Dagster's site before publishing.
Agentic AI
AI steps draw credits too, same as everything else.
Dagster
Every asset an AI step materializes, and every op it executes, draws down the same credit meter as regular code; Dagster doesn't separate AI-driven work from anything else for billing or cost visibility.
Unmeshed
Unmeshed keeps code steps free of charge and prices only the AI steps individually, showing the cost of each one live as the workflow runs, so AI usage stays visible on its own rather than blending into a shared step count.
Cost and infrastructure
Hybrid still means half of it isn't yours.
Dagster
Dagster's hybrid deployment lets you run compute in your own infrastructure, but the scheduler, UI, and metadata still live in Dagster's hosted control plane. Full independence means running the open-source project directly, without Dagster+, which also means giving up governed access control, SSO, audit logs, and AI features entirely.
Unmeshed
Unmeshed self-hosts the whole platform, control plane included, without trading away governance to do it.
Feature comparison
Dagster makes you plan for this. Unmeshed already did.
The stuff you won't have to build yourself.
Pricing predictability
One number, not a credit per step.
Unmeshed
Workflow executions are counted flat, no matter how many steps a workflow runs or how much compute it uses.
Dagster
A credit is charged for every asset materialized and every op executed, plus serverless compute billed separately by the minute.
Self-hosting
The control plane included, not left behind.
Unmeshed
Self-host on your own VM or Kubernetes cluster, or use Unmeshed Cloud, control plane included. On-premises deployment is included on the Enterprise plan.
Dagster
Hybrid deployment runs your compute on your own infrastructure, but the scheduler, UI, and metadata still live in Dagster's hosted control plane.
Governance
Some access control before the top tier.
Unmeshed
Role-based access control is available as a baseline platform capability, not absent on the entry plan.
Dagster
The Solo plan has no role-based access control at all, and SSO, SAML, SCIM, and audit logs are all reserved for the custom Pro plan, nothing in between.
Workflow definitions
Define it your way, not just in Python.
Unmeshed
Workflows are defined in JSON, YAML, or code, whichever fits the team building it.
Dagster
Assets and pipelines are defined in Python only, structured as software-defined assets rather than a JSON or YAML option.
Code sample
Just functions. No credit meter to watch.
Workers are plain functions. Register them, start a process, and the platform takes care of the durable execution machinery, no per-step credit meter running in the background.
Retries, state, and scheduling come built in, so there's no extra worker choreography to hand roll.
from unmeshed.sdk.configs.client_config import ClientConfig
from unmeshed.sdk.unmeshed_client import UnmeshedClient
from unmeshed.sdk.decorators.worker_function import worker_function
client_config = ClientConfig()
client_config.set_client_id("your-client-id")
client_config.set_auth_token("your-auth-token")
client = UnmeshedClient(client_config)
@worker_function(name="charge-order", namespace="default", max_in_progress=100)
def charge_order(input: OrderInput) -> ChargeResult:
# your business logic — the engine handles
# state, retries, and observability
return charge_card(input)
client.start()Frequently asked questions
Still have questions? Talk to us.