ibis-framework

What it is and what it does

Ibis provides a unified Python API for dataframe operations that compiles to SQL or native backend code, letting you write once and run against multiple execution engines. The library supports both interactive mode for exploration and lazy evaluation for performance, and lets you mix Python and SQL code seamlessly. You can iterate locally, then deploy the same code against different backends by changing a single line.

Runtime dependencies are minimal—parsy, sqlglot, toolz, python-dateutil, typing-extensions, tzdata, and atpublic—and the package itself is pure Python with no compiled dependencies. The library has been actively maintained since its first release in 2015.

Use it for:

• Develop data pipelines locally, then deploy to different backends without rewriting logic.
• Build exploratory data analysis notebooks that work across multiple backends without backend-specific code.
• Compose complex SQL queries programmatically using Python expressions instead of string concatenation.
• Migrate data workloads between backends by changing backend configuration.
• Mix SQL subqueries with Python dataframe operations for hybrid workflows.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Ibis is a portable Python dataframe library that compiles dataframe expressions into SQL or native operations, letting you use a single API across more than 20 backends.

Yes. Ibis is actively maintained, production-stable, has no known vulnerabilities, and solves a real problem: backend portability. The low install friction and permissive Apache-2.0 license make it a safe choice. Install it if you work with multiple data backends or want to avoid vendor lock-in; skip it if you're committed to a single backend and don't need portability.

Install

ibis-framework on PyPI

pip install ibis-framework

uv add ibis-framework

poetry add ibis-framework

Installing ibis-framework

Before you install

Low friction: pure Python wheel with 7 lightweight runtime dependencies. Actively maintained with recent commits and 6624 GitHub stars. Requires Python 3.10 or later.

License in practice

Apache-2.0 permissive license allows commercial use, modification, and redistribution with minimal restrictions—suitable for most projects.

Quickstart

pip install 'ibis-framework[duckdb]'

import ibis
ibis.options.interactive = True
t = ibis.examples.penguins.fetch()
result = t.group_by('species').agg(count=t.count())

Requires Python 3.10 or later. Optional backend packages must be installed separately for non-default execution.

Verify before relying

• Performance characteristics and query optimization strategies compared to direct backend usage.
• Completeness of SQL feature coverage across all supported backends.
• Memory overhead of lazy expression compilation and deferred execution.

Package facts

Evidence: ibis_framework-12.0.0-py3-none-any.whl

Keywords: bigquery, clickhouse, database, datafusion, duckdb, expressions, impala, mssql, mysql, pandas, polars, postgresql, pyarrow, pyspark, singlestoredb, snowflake, sql, sqlite, trino

Tags

More Scientific/Engineering packages