LLM Providers

LLM plugins power the agent’s dialogue and reasoning.

Set your API key in .env

Create a .env file (same directory where you run the worker) and add your provider key:

# .env
OPENAI_API_KEY=...

How to use an LLM plugin

Worker defaults

from dotenv import load_dotenv
from siphon.agent import Agent
from siphon.plugins import openai

load_dotenv()

agent = Agent(
    agent_name="Receptionist",
    llm=openai.LLM(),
    system_instructions="You are a helpful receptionist.",
)

if __name__ == "__main__":
    agent.dev()

Per-call overrides

You can also pass an LLM object into Dispatch(...) or Call(...) for per-call overrides.

• Agents Dynamic Configuration

Supported LLM providers

• OpenAI
• Anthropic
• Groq
• Gemini
• OpenRouter
• xAI
• DeepSeek
• Ollama
• Perplexity
• Together
• Mistral
• Cerebras