Integration · Multi-agent frameworkOpenAI-compatible · one key · live model catalog

Run the EveryAPI gateway inside CrewAI

CrewAI can construct an LLM with an OpenAI-compatible base URL, key, and model name. Point that LLM at EveryAPI and assign it to the agents that should use it.

Get your EveryAPI API key →Visit the CrewAI website ↗

01Setup code

One base URL line points CrewAI at EveryAPI

Default global Base URL / China-first Base URL: https://api.everyapi.ai/v1 / https://api-cn.everyapi.ai/v1

app.py

from crewai import Agent, Crew, Task, LLM
 
# CrewAI's LLM abstraction routes through LiteLLM — OpenAI-compatible prefix "openai/"
llm = LLM(
    model="openai/claude-sonnet-5",
    base_url="https://api.everyapi.ai/v1",
    api_key="sk-everyapi-...",
)
 
researcher = Agent(
    role="Senior Research Analyst",
    goal="Find signal in the latest data",
    backstory="...",
    llm=llm,
)

02Steps

From zero to your first call

Install crewai
Install the CrewAI version used by your project and consult that version’s LLM configuration reference.
Set base_url + api_key on the LLM instance
crewai.LLM(model="openai/...", base_url=..., api_key=...). Note the openai/ prefix on the model name, which tells LiteLLM to use the OpenAI-compatible path.
Run with crew.kickoff()
Run a small crew and verify its requests first. Agent tools, structured responses, and other features still depend on the chosen model and CrewAI version.

04Other integrations

EveryAPI plugs into these too

Stateful agent orchestration in the LangChain family

LangGraph →Use an EveryAPI-backed ChatOpenAI client inside a LangGraph workflow.

TypeScript AI agent framework

Mastra →Connect a current Mastra project to EveryAPI through a supported OpenAI-compatible provider.

Agent operating system written in Rust

LibreFang →Official integration: run agents with LibreFang and power them through EveryAPI.

LLM application framework

LangChain →Use EveryAPI as a unified upstream in your LangChain apps.

HLLM Observability

Helicone →Optionally place Helicone proxy mode in front of EveryAPI after reviewing the privacy and reliability trade-offs.

LSelf-hosted open-source chat UI

LibreChat →Add EveryAPI as a custom endpoint in a current LibreChat deployment.View all integrations →