vLLM Benchmark
STDIOInteractive tool for benchmarking vLLM performance and capabilities.
Interactive tool for benchmarking vLLM performance and capabilities.
This is proof of concept on how to use MCP to interactively benchmark vLLM.
We are not new to benchmarking, read our blog:
This is just an exploration of possibilities with MCP.
{
"mcpServers": {
"mcp-vllm": {
"command": "uv",
"args": [
"run",
"/Path/TO/mcp-vllm-benchmarking-tool/server.py"
]
}
}
}
Then you can prompt for example like this:
Do a vllm benchmark for this endpoint: http://10.0.101.39:8888
benchmark the following model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
run the benchmark 3 times with each 32 num prompts, then compare the results, but ignore the first iteration as that is just a warmup.