ClawBench

What is ClawBench?

ClawBench is a Browser Agents tool. Open-source benchmark for browser AI agents on daily tasks. ClawBench is free to use. It is available on macOS. It is aimed at developers and researchers.

ClawBench is an open-source benchmark designed for evaluating browser AI agents on daily tasks. It provides tools for configuring models, running tasks, and analyzing results. The platform supports developers and researchers in testing and improving AI agent performance.

Built for developers and researchers.

Key capabilities

• Agents
• Summarization

Features

• Quick Start

• Install

• Configure Models

• Other Browser Runtimes and Frameworks

• Datasets

• Awesome Works Using ClawBench

• Results

• Reproduce the Leaderboard

• Path A

  Re-runs the agent and scores the performance.

Use cases

• Configure models for task execution
• Run your first task using the benchmark
• Select a harness for task frameworks
• Access datasets for testing and evaluation
• Analyze results and performance metrics
• Reproduce leaderboard results for validation

Pros & cons

Pricing

Free.

Developer & API

FAQ