dbt-labs/metricflow

Health v2 · maintenance only

Adoption not part of the score

1768 stars · 199 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

MetricFlow is a Python semantic layer library from dbt Labs that lets you define, build, and maintain business metrics in code. It compiles metric definitions into optimized, engine-specific SQL, handling multi-hop joins, ratio/expression/cumulative metrics, and time-granularity aggregation.

Use cases

• define business metrics as code in yaml
• generate consistent sql queries from metric definitions
• build a semantic layer on top of a data warehouse
• query metrics with dimensions and time granularities
• avoid duplicated metric logic across bi tools
• compute ratio and cumulative metrics from warehouse data

When to choose

• you use dbt and want centralized, version-controlled metric definitions
• you need consistent metric logic across multiple BI tools and consumers
• you want automated SQL generation with multi-hop joins and complex metric types
• you target Snowflake, BigQuery, Databricks, Postgres, or Redshift

When to avoid

• you need a standalone BI dashboard or visualization tool
• you don't use dbt or a SQL data warehouse
• you need real-time streaming metrics rather than warehouse queries
• you want a no-code metrics interface for non-technical users

Facets

library · maturity active

data-science analytics cli developer-tools analytics big-data developer-tools python cli cross-platform semantic-layer metrics-as-code sql-generation dbt business-intelligence data-modeling data-engineering

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dbt-labs/metricflow")