ASSERT Documentation Overview

Welcome to ASSERT!

Install

Install the package and run your first evaluation in minutes.

Prerequisite: Python 3.11+

pip install assert_ai

Start here

Begin with the essential overview and first-run walkthrough.

  • Getting Started: Run your first evaluation with a canonical example
  • Concepts: Understand the concepts of the ASSERT evaluation framework, its pipeline mental model and key terminology.

How-to guides

Step-by-step guides for common evaluation tasks.

  • Create an Evaluation: Build a new eval config using the interactive assistant or manual YAML.
  • Results Guide: Interpret scores, metrics, and judge evidence from completed runs.
  • Local Viewer: Explore suites, runs, transcripts, and metrics in the local web UI.
  • Troubleshooting: Resolve common setup, runtime, and provider integration issues.

Configuration

Reference docs for writing and tuning eval configuration files.

  • Config Overview: Learn the structure and components of an eval config YAML file required for running evaluations.
  • Config Schema: Reference every supported YAML field, type, and default behavior.
  • Best Practices and Limitations: Avoid common pitfalls and understand current pipeline limitations.

CLI

Command reference for creating, running, and inspecting evaluations.

  • CLI Overview: Learn the core CLI workflow for initializing, running, and comparing evaluations.
  • CLI Commands: Browse command syntax, options, and examples for each CLI command group.

Targets

Choose the right target integration path for your system.

Additional docs and examples for deeper exploration.