Dagster is an open-source data orchestration platform for asset-based pipelines. See pricing, features, and its 2026 acquisition by Prefect.
Category
Automation
Pricing
Freemium / Open Source, from Free (self-hosted); Dagster+ from $120/month
Verified
Not yet
Last updated
July 18, 2026
Founded
2018
Headquarters
San Francisco, California, United States
Web AppFree TrialAPIOpen SourceAISelf-Hosted
Overview
Dagster is an open-source data orchestration platform, created by Nick Schrock and originally built by the company Elementl (later renamed Dagster Labs), founded in 2018 in San Francisco.
It organizes pipelines around software-defined assets rather than plain tasks, giving teams built-in lineage and observability, and in July 2026 Dagster Labs was acquired by orchestration rival Prefect, which committed to keeping both products and their pricing unchanged.
Key Features
Dagster pipelines are authored in Python and integrate with dbt, Spark, Snowflake, BigQuery, Databricks, and common cloud storage, with growing support for orchestrating AI/LLM and agentic workflows alongside traditional ETL.
The open-source core (Apache 2.0) includes the DAG engine, scheduler, sensors, and a local web UI, while the managed Dagster+ product adds a hosted control plane, catalog search, cost tracking, RBAC, audit logging, and uptime SLAs.
Pricing
Self-hosting the open-source core is free. Dagster+ Solo costs $120/month with 7,500 credits, and Starter costs $1,200/month with 30,000 credits, both under the pricing structure effective May 1, 2026.
An Enterprise tier is available on a custom quote basis with unlimited code locations and deployments, and a 30-day free trial with unlimited users is offered for teams evaluating Dagster+.
Key Features
Software-defined assets — Declares data assets and their dependencies rather than raw tasks, giving automatic lineage and dependency graphs.
Python-native pipeline authoring — Pipelines, schedules, and sensors are defined in Python with strong local testability.
Modern data stack integrations — Native integrations with dbt, Spark, Snowflake, BigQuery, Databricks, and major cloud object stores.
AI and agent pipeline support — Increasingly used to orchestrate RAG pipelines, LLM workflows, and agent-based data tasks.
Hybrid or serverless execution — Dagster+ can run customer code in the customer's own cloud (hybrid) or fully managed (serverless).
Built-in data catalog — Dagster+ provides catalog search across pipelines, assets, and metadata for platform-wide visibility.
Cost and usage tracking — Dagster+ tracks credit consumption and compute cost per asset, pipeline, or team.
RBAC, audit logging, and SLAs — Enterprise-grade governance features available on paid Dagster+ plans.
Pros & Cons
Pros
Asset-based model gives built-in lineage and observability other orchestrators require add-ons for.
Strong local development and testing experience for Python-based pipelines.
Free, fully-featured open-source core with no artificial feature gating.
Post-acquisition, Prefect has committed to continued investment and unchanged pricing.
Cons
The asset-centric abstraction has a steeper learning curve than simpler task-based schedulers.
Dagster+ credit-based pricing can be harder to forecast than flat per-seat pricing at scale.
Enterprise pricing is not publicly listed and requires a sales conversation.
Recent acquisition by Prefect introduces some uncertainty about long-term product direction, despite stated commitments.
Pricing
Open Source Free N/A
Dagster+ Solo $120/month monthly (pay-as-you-go, 7,500 credits included)
Yes, the open-source core of Dagster is free and Apache 2.0 licensed and can be self-hosted indefinitely. Dagster+ is a paid managed offering on top of it.
What is Dagster+?
Dagster+ (formerly Dagster Cloud) is Dagster Labs' managed control plane, adding a hosted UI, catalog search, cost tracking, RBAC, audit logging, and SLAs, priced from $120/month for the Solo plan.
Who created Dagster?
Dagster was created by Nick Schrock, a former co-creator of GraphQL, who founded the company (originally Elementl, later renamed Dagster Labs) in 2018.
Did Dagster get acquired?
Yes. Prefect announced on July 13, 2026 that it was acquiring Dagster Labs, with both open-source Dagster and Dagster+ continuing to be maintained and priced as before under Prefect.
How is Dagster different from Airflow?
Dagster centers on software-defined assets with built-in lineage and testability, while Airflow is a more task-centric scheduler; Dagster was explicitly designed as a next-generation alternative to Airflow.
Can Dagster orchestrate AI/LLM workflows?
Yes, Dagster is increasingly used to orchestrate RAG pipelines, LLM calls, and agentic workflows alongside traditional data engineering pipelines.