Skip to content

Dataset · TrangBui

Dynamic multi-agent routing benchmark

TrangBui/RoleRouteBench

English benchmark with a reproducible environment for evaluating task routing between AI agents.

What's inside

The dataset represents a reproducible multi-domain environment (formerly known as RoutingBench V4-beta) for evaluating dynamic routing in multi-agent systems. It contains between 1,000 and 10,000 English records in tabular and text formats, and includes a fixed execution environment and package identifiers for clean algorithm comparison.

What you can do with it

  • Evaluating and comparing dynamic routing algorithms in multi-agent systems
  • Training classifiers to distribute requests among specialized agent roles
Tasks
Text classification
Languages
English
Size
1K–10K rows
Downloads this month
781
Likes
2

How to load it

from datasets import load_dataset

ds = load_dataset("TrangBui/RoleRouteBench")