ibis-project/ibis

Health v2 · maintenance only

Adoption not part of the score

6643 stars · 755 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Ibis is a portable Python dataframe library that provides a single lazy dataframe API across more than 20 execution backends including DuckDB, Polars, Postgres, BigQuery, and Spark. It compiles Python dataframe expressions to SQL or native backend code, letting users iterate locally and deploy remotely by changing one line.

Use cases

• query databases with a python dataframe api instead of writing sql
• analyze parquet files locally with duckdb
• write analytics code once and run it on snowflake, bigquery, or postgres
• mix python dataframe code with sql in one workflow
• prototype data pipelines locally then deploy to a warehouse
• explore datasets interactively in a notebook
• migrate pandas workflows to a scalable sql engine

When to choose

• you want one dataframe API that works across many SQL engines and local engines
• you need lazy evaluation and deferred execution for large datasets
• you want to bridge python data science code with SQL-based data engineering
• you iterate on small local data but production runs on a warehouse or Spark

When to avoid

• you only need simple in-memory pandas operations with no backend portability
• you need fine-grained control over engine-specific SQL features
• your team already standardizes on a single engine's native API

Facets

library · maturity active

data-science etl database parser data-science databases analytics big-data python cross-platform dataframe lazy-evaluation sql-generation duckdb pandas polars pyspark bigquery snowflake portable-api data-exploration sql data-engineering

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ibis-project/ibis")