Python

2.0x faster throughput.
Production HTTP transport.

ZeroMCP vs zeromcp-mrexodia — HTTP benchmarks.

Throughput
3.53K req/s ZeroMCP
vs.
1.80K req/s mrexodia
Memory
30 MB ZeroMCP
vs.
13 MB mrexodia

HTTP Performance — Head to Head

Same hello tool. Same methodology. 5-minute sustained load in Docker. Starlette for ZeroMCP, built-in transport for zeromcp-mrexodia.

req/s p50 p99 Memory CPU
ZeroMCP (Starlette) 3.53K 0.27ms 0.57ms 30 MB 0.28%
zeromcp-mrexodia 1.80K 0.45ms 1.24ms 13 MB 0.04%

The tradeoff

Choose mrexodia

If memory footprint is your top priority. 13 MB is hard to beat. Great for resource-constrained environments.

  • 13 MB memory footprint
  • Lightweight built-in transport
  • Great for resource-constrained environments
Choose ZeroMCP

If you need production throughput. 2.0x faster on Starlette with enforced sandboxing and credential injection.

  • 0 dependencies
  • File-based tools — drop a .py file, it's live
  • Starlette + FastAPI + Flask
  • Built-in sandbox with enforced permissions
  • 3.53K req/s on Starlette

Drop a .py file. It's an MCP tool.