Integrationv1.0

CrewAI + Meridian

Wire your CrewAI agents directly into Meridian's runtime. Use any ChatOpenAI-compatible endpoint — OpenAI, Azure, Together, or your own vLLM instance — with full control over base URL and model selection.

ChatOpenAI Configuration

Set your OpenAI-compatible endpoint. Meridian passes these to the CrewAI LLM constructor at runtime. Stored locally in your browser.

Usage

Meridian injects the LLM into your CrewAI agent. No code changes needed.

from crewai import Agent
from meridian import MeridianLLM

llm = MeridianLLM.from_config()  # reads your saved settings

researcher = Agent(
    role="Senior Researcher",
    goal="Uncover cutting-edge developments",
    backstory="You are a veteran analyst.",
    llm=llm,
    verbose=True,
)