Connect to the remote VLM Run MCP Server and start building agentic workflows with visual AI in minutes.
Get your API key
$VLMRUN_API_KEY
). We’ll use this to authenticate your requests to the MCP server next.Ping the MCP server to test
Add the server to your MCP client
~/Library/Application Support/Claude/claude_desktop_config.json
%APPDATA%/Claude/claude_desktop_config.json
See docs for more details.Test the connection
put_image_url
, put_document_url
, detect_faces
, etc.Claude Desktop not showing MCP tools
npx
is installed and accessiblehttps://mcp.vlm.run/mcp/sse
with Bearer token authenticationOpenAI Responses API SDK not recognizing MCP server
Claude Pro (Web) integration issues
https://mcp.vlm.run/mcp/sse
with proper Bearer token authentication when adding the integrationCursor not showing MCP tools
~/.cursor/mcp.json
file exists in your project roothttps://mcp.vlm.run/mcp/sse
with Bearer token authenticationnpx
is installed and accessible from your system PATHConnection timeout or server unreachable
https://mcp.vlm.run/mcp/sse
is accessible (note: authentication is required via Bearer token)Tool not found or extraction errors
list_hub_domains
first to verify connectionextract_image
not extract-image
)document.invoice
not invoice
)