Deep Dives

This section contains in-depth guides for working with Scale Gymnasium. Choose a topic based on what you want to learn.

Guides

Complete walkthrough of the Gymnasium Web interface

Creating effective tasks for agent evaluation

State checks, log checks, and rubric evaluation

Pre-configured datasets for environments

Common issues and solutions

For environment-specific information, visit the Environments section:

Calendr, Cloudfile, Shopora, Pandora’s Inbox

Windows, Ubuntu, macOS VMs

45+ MCP servers with 300+ tools

⌘I