crewplus 0.2.25

Base services for CrewPlus AI applications

CrewPlus

CrewPlus provides the foundational services and core components for building advanced AI applications. It is the heart of the CrewPlus ecosystem, designed for scalability, extensibility, and seamless integration.

Overview

This repository, crewplus-base, contains the core crewplus Python package. It includes essential building blocks for interacting with large language models, managing vector databases, and handling application configuration. Whether you are building a simple chatbot or a complex multi-agent system, CrewPlus offers the robust foundation you need.

The CrewPlus Ecosystem

CrewPlus is designed as a modular and extensible ecosystem of packages. This allows you to adopt only the components you need for your specific use case.

• crewplus (This package): The core package containing foundational services for chat, model load balancing, and vector stores.
• crewplus-agents: An extension for creating and managing autonomous AI agents.
• crewplus-ingestion: Provides robust pipelines for knowledge ingestion and data processing.
• crewplus-memory: Provides agent memory services for Crewplus AI Agents.
• crewplus-integrations: A collection of third-party integrations to connect CrewPlus with other services and platforms.

Features

• Chat Services: A unified interface for interacting with various chat models (e.g., GeminiChatModel).
• Model Load Balancer: Intelligently distribute requests across multiple LLM endpoints.
• Vector DB Services: working with popular vector stores (e.g. Milvus, Zilliz Cloud) for retrieval-augmented generation (RAG) and agent memory.

Documentation

For detailed guides and API references, please see the docs/ folder.

• GeminiChatModel Documentation: A comprehensive guide to using the GeminiChatModel for text, image, and video understanding.

Installation

To install the core crewplus package, run the following command:

pip install crewplus

Getting Started

Here is a simple example of how to use the GeminiChatModel to start a conversation with an AI model.

# main.py
from crewplus.services import GeminiChatModel

# Initialize the llm (API keys are typically handled by the configuration module)
llm = GeminiChatModel(google_api_key="your-google-api-key")

# Start a conversation
response = llm.chat("Hello, what is CrewPlus?")

print(response)

Project Structure

The crewplus-base repository is organized to separate core logic, tests, and documentation.

crewplus-base/                    # GitHub repo name
├── pyproject.toml
├── README.md
├── LICENSE
├── CHANGELOG.md
├── crewplus/                 # PyPI package name
│   └──  __init__.py
│   └──  services/
│       └──  __init__.py
│       └──  gemini_chat_model.py
│       └──  model_load_balancer.py
│       └──  ...
│   └──  vectorstores/milvus
│       └──  __init__.py
│       └──  schema_milvus.py
│       └──  vdb_service.py
│   └──  core/
│       └──  __init__.py
│       └──  config.py
│       └──  ...
├── tests/
│   └── ...
├── docs/
│   └── ...
└── notebooks/
    └── ...

Deploy to PyPI

Clean Previous Build Artifacts: Remove the dist/, build/, and *.egg-info/ directories to ensure that no old files are included in the new build.

rm -rf dist build *.egg-info

install deployment tool

pip install twine

build package

python -m build

deploy to TestPyPI (Test first)

python -m twine upload --repository testpypi dist/*

install from TestPyPI

pip install -i https://test.pypi.org/simple/ crewplus

Deploy to official PyPI

python -m twine upload dist/*