BigHugger
GH Repository · georgia-tech-db

evadb

Database system for AI-powered apps

stars
2,674
30-day movement
starts with the next reading
Related entries
60
Connections
1
Pythonhacktoberfestagentgpt-4pythonllmevaauto-gptagent-frameworkdatabasedata-analysisgpt4allailabelingobject-detectionvideo-analyticsservingchatgpthuggingfacelangchain

EvaDB is a database system for building AI-powered applications, developed by Georgia Tech's DB group. Based on its topics, it combines database querying with video analytics, labeling, object detection, and LLM integrations such as ChatGPT, LangChain, GPT-4, and Hugging Face models.

It lets you query and analyze data—especially video—with AI models and LLMs through a database interface rather than gluing separate tools together.

Use it to

  • Run AI-powered queries over video data
  • Perform object detection and labeling through SQL-like access
  • Integrate LLMs like GPT-4 or LangChain into data analysis pipelines
  • Build agentic applications backed by a database

For Developers building AI-powered data and video analytics apps

Role
agent-framework
Language
Python
Licence
Apache-2.0
Forks
258
Open issues
78
Last push
2024-05-17
Latest release
v0.0.3 · 2022-07-31
topicsdatabasevideo-analyticsllmagent-frameworkobject-detectiondata-analysis