pdf4vllm MCP
A community-driven MCP server that reads PDFs for vision LLMs. It auto-detects text corruption and switches to image mode, preserves reading order, prevents context overflow via page limits, and filters unnecessary images like logos and headers.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Find PDF files using glob patterns
- Extract PDF content as ordered text, table, and image blocks
- Search text within PDFs using pdfgrep
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
pdf4vllm-mcp 1.0.2
{
"version": "1.0.2",
"transport": "stdio",
"identifier": "pdf4vllm-mcp",
"registryType": "pypi"
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"pdf4vllm-mcp": {
"command": "uvx",
"args": [
"pdf4vllm-mcp==1.0.2"
]
}
}
}Cline
{
"mcpServers": {
"pdf4vllm-mcp": {
"command": "uvx",
"args": [
"pdf4vllm-mcp==1.0.2"
]
}
}
}VS Code
{
"servers": {
"pdf4vllm-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"pdf4vllm-mcp==1.0.2"
]
}
}
}Claude Desktop
{
"mcpServers": {
"pdf4vllm-mcp": {
"command": "uvx",
"args": [
"pdf4vllm-mcp==1.0.2"
]
}
}
}Let your agent find its tools
Find servers like pdf4vllm MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.