EveryMCP

Indexed MCP Listing

vLLM MCP

High-throughput open-source LLM serving with PagedAttention, exposed via MCP.

AI & ML ToolsSelf-hostedAIIndexed source: vllm-project
Before you install: EveryMCP has not security-audited this listing or independently verified its publisher. Review the linked repository, requested permissions, deployment model, and official-registry record where applicable. How the catalog works →
Listed author
vllm-project
Repository
https://github.com/vllm-project/vllm

Installation reference

Deploy vLLM with your GPU setup; mount the OpenAI-compatible endpoint via MCP.

Treat this as catalog metadata, not an audited command. Confirm current instructions in the linked sources before running it.

Use Cases

  • • Self-hosted LLM inference
  • • Batch generation
  • • Streaming completions

Tags

llm-servingopen-sourcegpuinference

Choose a starting stack

The free planner offers three-server shortlists for five common goals. It does not automatically include vLLM MCP or confirm compatibility with your environment.

Build a free starting stack

Rolling out your first MCP workflow? The $49 starter kit provides a selection matrix, permission record, client example, and acceptance/rollback checklist. It does not certify vLLM MCP or provide managed setup.

Review the rollout packet — $49

For complex enterprise requirements, share consulting context.