Skip to main content

Chat with your data by leveraging the power of LLMs and vector databases

Project description

ContextQA logo

ContextQA title

Chat with your data by leveraging the power of LLMs and vector databases

contextqa latest version Supported Python versions node version vue version


ContextQA is a modern utility that provides a ready-to-use LLM-powered application. It is built on top of giants such as FastAPI, LangChain, and Hugging Face.

Key features include:

  • Regular chat supporting knowledge expansion via internet access
  • Conversational QA with relevant sources
  • Streaming responses
  • Ingestion of data sources used in QA sessions
  • Data sources management
  • LLM settings: Configure parameters such as provider, model, temperature, etc. Currently, the supported providers are OpenAI and Google
  • Vector DB settings. Adjust parameters such as engine, chunk size, chunk overlap, etc. Currently, the supported engines are ChromaDB and Pinecone
  • Other settings: Choose embedded or external LLM memory (Redis), media directory, database credentials, etc.

Installation

pip install contextqa

Usage

On installation contextqa provides a CLI tool

contextqa init

Check out the available parameters by running the following command

contextqa init --help

Example

Run it

$ contextqa init

2024-08-28 01:00:39,586 - INFO - Using SQLite
2024-08-28 01:00:47,850 - INFO - Use pytorch device_name: cpu
2024-08-28 01:00:47,850 - INFO - Load pretrained SentenceTransformer: sentence-transformers/all-mpnet-base-v2
INFO:     Started server process [20658]
INFO:     Waiting for application startup.
2024-08-28 01:00:47,850 - INFO - Running initial migrations...
2024-08-28 01:00:47,853 - INFO - Context impl SQLiteImpl.
2024-08-28 01:00:47,855 - INFO - Will assume non-transactional DDL.
2024-08-28 01:00:47,860 - INFO - Running upgrade  -> 0bb7d192c063, Initial migration
2024-08-28 01:00:47,862 - INFO - Running upgrade 0bb7d192c063 -> b7d862d599fe, Support for store types and related indexes
2024-08-28 01:00:47,864 - INFO - Running upgrade b7d862d599fe -> 3058bf204a05, unique index name
INFO:     Application startup complete.
INFO:     Uvicorn running on http://localhost:8080 (Press CTRL+C to quit)

Check it

Open your browser at http://localhost:8080. You will see the initialization stepper which will guide you through the initial configurations

init config

Or the main contextqa view - If the initial configuration has already been set

main view

Guideline

For detailed usage instructions, please refer to the usage guidelines.

Contributing

We welcome contributions to ContextQA! To get started, please refer to our CONTRIBUTING.md file for guidelines on how to contribute. Your feedback and contributions help us improve and enhance the project. Thank you for your interest in contributing!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

contextqa-2.0.5.tar.gz (2.7 MB view details)

Uploaded Source

Built Distribution

contextqa-2.0.5-py3-none-any.whl (2.7 MB view details)

Uploaded Python 3

File details

Details for the file contextqa-2.0.5.tar.gz.

File metadata

  • Download URL: contextqa-2.0.5.tar.gz
  • Upload date:
  • Size: 2.7 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.11.9

File hashes

Hashes for contextqa-2.0.5.tar.gz
Algorithm Hash digest
SHA256 27560f19c1e5d5df292de94f3046e30c39347bd826adf6a06690d0e3de218cbb
MD5 1f440936f9e859081926c30178b08a29
BLAKE2b-256 d01f4dc1f10dfcc799e21b242b79e9ec4e11804c31a25458395d753a40afb772

See more details on using hashes here.

File details

Details for the file contextqa-2.0.5-py3-none-any.whl.

File metadata

  • Download URL: contextqa-2.0.5-py3-none-any.whl
  • Upload date:
  • Size: 2.7 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.11.9

File hashes

Hashes for contextqa-2.0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 61b02c688c7d5bbfc0534beab87802b6b44596acd28a8d24aab95a107034e811
MD5 7fc629aa4bad317cdaa2e58a8ce11a7d
BLAKE2b-256 509fa4264bf2c6763c21243ee685ad35d26cce6cd999f9341a6c16a61bdc490a

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page