TaskingAI is an open source platform for AI-native application development with a Python-based server, Docker-based deployment, and a client SDK. It provides tools for building AI agents with RAG and multi-model integrations.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
TaskingAI is an open source platform for AI-native application development.
How it works
The project provides a backend that integrates hundreds of LLM models, tools, RAG systems, and agents, accessible via unified APIs. It supports both server-side AI logic and client-side development, with a Docker-based deployment path and a Python client SDK for programmatic interaction.
Getting started
Quickstart with Docker Prerequisites: Docker and Docker Compose, Git, Python (>=3.8) for the client SDK. Installation:
git clone https://github.com/taskingai/taskingai.git
cd taskingai
Inside the repository, go to the docker directory:
cd docker
- Copy environment file:
cp .env.example .env
- Edit the .env file to set required configurations.
- Start services:
docker-compose -p taskingai --env-file .env up -d
Access the console at http://localhost:8080 with default credentials: admin / TaskingAI321. Upgrade:
git pull origin master
Then:
cd docker
docker-compose -p taskingai down
docker-compose -p taskingai pull
docker-compose -p taskingai --env-file .env up -d
Client SDK usage example is provided in the README under the Docker section, starting with:
pip install taskingai
Example Python code snippet demonstrates initializing the client and performing actions like creating an assistant, creating a chat, sending a message, and generating a response.
Documentation is available at https://docs.tasking.ai.
Recent releases
Latest releases include v0.3.0 (2024-06-03) with updates to Playground Markdown rendering and image uploads, and prior versions v0.2.2 (2024-04-12), v0.2.0 (2024-03-12), v0.1.3 (2024-02-06), and v0.1.2 (2024-02-02).
Traction
Stars: 5396 (as per repository facts). Open issues: 46. Language: Python. License: Apache-2.0.
Caveats
License: TaskingAI Open Source License. Created: 2024-01-08. Last push: 2024-12-02.
Behind the repo
Not provided in the README excerpt.