Run the EveryAPI gateway inside Mastra

One base URL line points Mastra at EveryAPI

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

From zero to your first call

1. 01
   Install Mastra

   Create or update a Mastra project using its current installation guide and record the package versions in your lockfile.

2. 02
   Point createOpenAI at EveryAPI

   Create an OpenAI-compatible provider using the API supported by your installed AI SDK and Mastra versions, with the EveryAPI base URL and key.

3. 03
   Pass the provider instance to the Agent

   Pass a model from that provider to an agent, using an ID available in your EveryAPI account catalog.

4. 04
   Test the features your workflow uses

   Agent, workflow, memory, RAG, structured-output, and tool behavior can depend on separate services, package versions, and model capabilities. Verify each path rather than assuming that changing the provider is transparent.

EveryAPI plugs into these too

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.

JS / TypeScript LLM framework

Vercel AI SDK →Route your Vercel AI SDK app through the EveryAPI gateway.

Python LLM framework

LlamaIndex →Route your LlamaIndex RAG / agent through the EveryAPI gateway.

Self-improving autonomous agent (Nous Research)

Hermes Agent →Configure the current Hermes Agent release to use EveryAPI through its documented custom-provider path.

Cross-platform AI assistant (the lobster way 🦞)

OpenClaw →Connect an installed OpenClaw release to EveryAPI using its current provider schema.View all integrations →