/

mirrored a minute ago

Benchmark Card Files and versions Leaderboard

Badge

Hub
Contact

Taskbench

Overview

This is a benchmark repository for Taskbench, part of the BenchFlow project. This benchmark is categorized as a agent benchmark.

Original Repository

This benchmark is based on main/taskbench.

BenchFlow Integration

This repository contains the necessary files to integrate with BenchFlow:

benchflow_interface.py: Interface for BenchFlow integration
README.md: This documentation file

Usage

Please refer to the BenchFlow documentation for usage instructions.

0

Tags

agent

multimodal

tool-calling

Information

Organization

BenchFlow

Release Date

April 18, 2025

Github

https://github.com/microsoft/JARVIS/tree/main/taskbench

Paper

https://arxiv.org/abs/2311.18760