A comprehensive, zero-dependency framework for Generative AI and Agentic AI
Project description
OmniGenAI
A comprehensive, zero-dependency Python framework for Generative AI (GenAI) and Agentic AI. Covers all features: text/image/video/audio generation, multimodal processing, fine-tuning, evaluation, tool calling, memory management, RAG, multi-agent orchestration, planning, and workflow automation. Pluggable with any model (OpenAI, Anthropic, Groq, local/Ollama, custom transformers, vision models) without vendor lock-in. Resolves all common issues: heavy dependencies, complex APIs, token limits, scalability, ease of use, and performance.
Features
- Zero Dependencies: Core logic relies purely on standard Python. Bring your own LLM client.
- Complete GenAI Coverage: Text, image, video, audio generation; multimodal; fine-tuning; evaluation metrics.
- Full Agentic AI: Tool calling, memory pruning, RAG, multi-agent, planning algorithms.
- Provider Agnostic: Swap seamlessly between any models.
- Scalable and Efficient: Modular design, async support, optimized memory.
- Easy to Use: Simple APIs, extensive docs, examples for everyone.
Quick Start
import os
from omnigenai import OmniAgent, tool, OpenAIProvider
from openai import OpenAI
@tool
def get_weather(location: str) -> str:
"""Gets the current weather."""
return f"Weather in {location}: 72°F and sunny."
client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
provider = OpenAIProvider(client)
agent = OmniAgent(
provider=provider,
model="gpt-4",
tools=[get_weather]
)
response = agent.run("What's the weather in Tokyo?")
print(response)
Installation
pip install omnigenai
# For extras: pip install omnigenai[all]
Publishing to PyPI
- Create a PyPI account at https://pypi.org/
- Create an API token at https://pypi.org/manage/account/token/
- Export credentials locally:
export TWINE_USERNAME="__token__"
export TWINE_PASSWORD="<your-token-here>"
- Build and publish:
chmod +x publish.sh
./publish.sh
Test upload first
chmod +x publish_testpypi.sh
./publish_testpypi.sh
When to bump the version
Update version in setup.py before every new release.
Documentation
See docs/ for full API reference, tutorials, and examples.
Contributing
MIT License. Contributions welcome!
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file omnigenai-0.1.0.tar.gz.
File metadata
- Download URL: omnigenai-0.1.0.tar.gz
- Upload date:
- Size: 4.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a75a8e6a384746c4cb230522fa40b54cd09e922a8af30af9a16b35c3b03e1ec2
|
|
| MD5 |
5086ea9cf1d0b69a501f9effb55e805b
|
|
| BLAKE2b-256 |
a1d914a953afae5944e3493418ae0b3e4c268dcde61726ff38132d6cbe4504b8
|
File details
Details for the file omnigenai-0.1.0-py3-none-any.whl.
File metadata
- Download URL: omnigenai-0.1.0-py3-none-any.whl
- Upload date:
- Size: 3.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
26b7da03f37d6dde508005bcdb7102686d87c7c4c4159a82bd8e20f034448076
|
|
| MD5 |
bf8940a69626391bf48c3fe909b563bf
|
|
| BLAKE2b-256 |
2daa1b772522091544b11f108ce4d0544f0c9bc7b136dea3d1d34456d0fece56
|