LLM Station

A unified, provider-agnostic agent framework for OpenAI, Google Gemini, and Anthropic Claude with tool integration.

Quick Start

Install

# Install from PyPI (recommended)
pip install llm-station

# Optional: Install with specific provider support
pip install llm-station[openai]     # OpenAI only
pip install llm-station[anthropic]  # Anthropic only  
pip install llm-station[google]     # Google only
pip install llm-station[all]        # All providers

# Development install
git clone https://github.com/your-repo/llm_station.git
cd llm_station
pip install -e .[dev]

Set Up API Keys

# Add to .env file
echo "OPENAI_API_KEY=your-openai-key" >> .env
echo "GEMINI_API_KEY=your-gemini-key" >> .env  
echo "ANTHROPIC_API_KEY=your-anthropic-key" >> .env

Use Any Provider

from llm_station import Agent
from dotenv import load_dotenv
import os

load_dotenv()

# Same interface, any provider
agent = Agent(
    provider="openai", # or "google", "anthropic"
    model="gpt-4o-mini",
    api_key=os.getenv("OPENAI_API_KEY")
)

# Basic chat
response = agent.generate("What is machine learning?")
print(response.content)

# Use tools with simple, memorable names
response = agent.generate(
    "Search for recent AI developments",
    tools=["search"]  # Auto-routes to best search provider
)

print(response.content)
if response.grounding_metadata:
    sources = response.grounding_metadata.get("sources", [])
    print(f"Found {len(sources)} sources")

# Multiple tools work together
response = agent.generate(
    "Research AI trends, analyze the data, and create a summary",
    tools=["search", "code", "json"]
)

Documentation

• OPENAI.md - OpenAI provider setup and tools
• GOOGLE.md - Google Gemini provider setup and tools
• CLAUDE.md - Anthropic Claude provider setup and tools