chdb-io/chdb

chDB is an in-process OLAP SQL Engine 🚀 powered by ClickHouse

What it solves

chDB allows you to run high-performance SQL analytical queries directly inside your application or browser without needing to install, configure, or manage a separate database server. It bridges the gap between familiar data analysis tools like pandas and the raw power of the ClickHouse analytical engine.

How it works

chDB is an in-process SQL OLAP engine. It embeds the ClickHouse engine directly into your process, allowing it to read and process data formats like Parquet, CSV, JSON, and Arrow with minimal memory overhead. It provides a "DataStore" API that mimics pandas syntax; when you perform operations, it lazily records them and compiles them into optimized SQL that executes using the ClickHouse engine. It also supports running in the browser via WebAssembly.

Who it’s for

It is for data scientists, developers, and analysts who want the performance of a professional-grade analytical database (ClickHouse) while working within the convenience of Python or browser-based workflows.

Highlights

  • Pandas-compatible: Use familiar pandas syntax to perform high-performance SQL operations.
  • Zero-install: Runs in-process or in the browser via WebAssembly, removing the need for a dedicated database server.
  • Format support: Directly query over 60 data formats, including Parquet, CSV, JSON, and Arrow.
  • AI-assisted: Includes built-in support to translate natural language prompts into executable SQL.
  • Efficient: Minimizes data copying between C++ and Python using memoryview.

Related

  • Project
    FreedomIntelligence/Awesome-AI4Med

    Awesome-AI4Med is a curated, structured repository of medical AI research resources, including models, datasets, and benchmarks, designed to help researchers and developers track field progress and accelerate innovation.

  • Project
    logpai/loghub

    Loghub is a free collection of real system log datasets for AI-driven log analytics research. It includes 20 datasets across six categories (distributed systems, supercomputers, OSes, mobile, server apps, standalone software), with metadata like line counts, sizes, and labeling status. Logs are intentionally unsanitized. Usage requires citing the Loghub paper and referencing the repo URL.

  • Project
    pytorch/audio

    A PyTorch-based audio library that provides tools for processing audio data for machine learning, featuring GPU acceleration and specialized audio transforms.

  • Project
    bytewax/awesome-public-real-time-datasets

    This project is a curated list of public real-time datasets and APIs, covering finance, transportation, weather, health, cybersecurity, sports, and more. It saves developers time by collecting live data feeds with categories and descriptions, including both free and paid options.