MCP Server

The Gateway runs a Model Context Protocol server on FastMCP. Agents such as Pydantic AI, LangChain, Mastra, the OpenAI Agents SDK, and Claude Code call the same pipeline as chat/completions. Authentication, billing, and metering match the REST API. Tiers are on Authentication and Rate Limits.

Property	Value
Server name	`vlmrun`
Endpoint	`POST https://gateway.vlm.run/mcp`
Transport	Streamable HTTP (FastMCP), stateless: no session id, no session affinity
Auth	Bearer token, or OAuth where the deployment enables it
Tools	`read_document`, `read_image`, `read_audio`, `read_video`, `list_models`, `get_model_info`, `get_completion`

Quickstart

Add the server

claude mcp add --transport http vlmrun https://gateway.vlm.run/mcp \
  --header "Authorization: Bearer $VLMRUN_API_KEY"

To click through the tools, run npx @modelcontextprotocol/inspector. Set the transport to Streamable HTTP and point it at the same URL.

Ask the agent to parse a document

Read https://storage.googleapis.com/vlm-data-public-prod/hub/examples/finance.sec-filings/tsla-8k.pdf
and tell me the date of the earliest event reported.

The agent picks read_document on its own.

Inspect the response

<document file_name="tsla-8k.pdf" file_hash="sha256:1b7f04c3…" file_bytes="48213" mimetype="application/pdf" npages="5" dpi="96">
<page id="0" format="markdown" width="612" height="792">
...
Date of report (Date of earliest event reported): October 2, 2024
</page>
</document>

Every read_* tool returns text like this, plus a typed ReadResult with the model that ran and its cost.

In this section

Page	Covers
Authentication	Bearer tokens, OAuth, and discovery endpoints
Connect an Agent	Client configs, and calling the server without a framework
Tools	`read_document`, `read_image`, `read_audio`, `read_video`
Reference	`list_models`, `get_model_info`, `get_completion`, the return shape, and errors

Chat Completions API

The REST surface behind the same pipeline.

List models

The live model catalog and capabilities.

Authentication

Bearer tokens, tiers, and anonymous access.

Supported Inputs

Content part types, document limits, and video knobs.

Overview

Platform

OpenAI Compatibility

TypeSafe Compatibility

MCP

Misc

Quickstart

In this section

Chat Completions API

List models

Authentication

Supported Inputs

​Quickstart

​In this section

​Related

Chat Completions API

List models

Authentication

Supported Inputs

Quickstart

In this section

Related