Overview
SWE-bench CLI is a command-line tool for interacting with the SWE-bench API. This tool allows you to:
- Submit model predictions for evaluation
- Retrieve evaluation reports
- Manage your evaluation runs
- Track your model's performance
All on the cloud!
Key Features
- Easy Submission: Submit your model's predictions with a single command
- Real-time Tracking: Monitor evaluation progress in real-time
- Run Management: Access and delete runs as needed
Quick Links
- Installation: Get started with installing the CLI
- Authentication: Set up your API key
- Quick Start: Submit your first predictions
- User Guide: Detailed guide on using the CLI