MCP-Bench

Emerging

1papers using it

2026first seen

MCP-Bench is a benchmark dataset that contains tasks designed to evaluate the execution feasibility of tool workflows in the context of compact language models and their ability to interact with live tool catalogs.

🔎 Find this dataset

Papers using MCP-Bench (1)

Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents2026