# Streamable HTTP and stdio: the two MCP transports

MCP has two standard transports. stdio connects an app to a server it starts on the same computer, through standard input and output. Streamable HTTP connects an app to a server over the network with HTTP POST, optionally streaming responses, and supports normal web authentication.

## Which to use

- **stdio** for [local servers](https://ai-powered.com/academy/glossary/remote-vs-local-mcp-server): no network, one app per server, and the security guidance recommends it for local servers because only the app that started the server can talk to it.
- **Streamable HTTP** for remote servers: one server for many users, with bearer tokens, API keys or OAuth.

## What about SSE?

The older HTTP+SSE transport is deprecated in the 2026-07-28 spec, with a year-long off-ramp. Some apps still accept it; Antigravity, for example, only takes Streamable HTTP for remote servers.

## Sources

Official documentation, checked on 19 November 2026:

- [MCP architecture overview (modelcontextprotocol.io)](https://modelcontextprotocol.io/docs/learn/architecture)
- [The 2026-07-28 specification release (MCP blog)](https://blog.modelcontextprotocol.io/posts/2026-07-28/)

## Related terms

- [Remote vs local MCP server](https://ai-powered.com/academy/glossary/remote-vs-local-mcp-server): A local MCP server is a program the AI app starts on your own computer and talks to over standard input and output (stdio). A remote MCP server runs on someone else's machine and the app reaches it over the internet by URL, using Streamable HTTP and usually a sign-in.
- [MCP host, client and server](https://ai-powered.com/academy/glossary/mcp-host-client-server): In MCP, the host is the AI application you use (such as Claude Desktop or VS Code), a client is the part of the host that keeps one connection to one server, and a server is the program that offers tools, resources and prompts. A host with three servers runs three clients.
- [MCP](https://ai-powered.com/academy/glossary/mcp): MCP (Model Context Protocol) is an open standard for connecting AI applications to outside tools and data. An AI app connects to small programs called MCP servers, asks each one what it offers, and calls those tools when a task needs them.
