Compare your AI models on the same examples.
Evaluates AI models on datasets and records experiments in Arize for performance comparisons.
When to use it
Use when asked to create or run experiments, compare models, analyze experiment results, or benchmark AI performance.
Give it a model evaluation request; it collects outputs and evaluations and creates an experiment in Arize.
Requires the ax CLI, version 0.14.0 or higher; version checks are instructed only after a command fails.
Requires authenticated access to Arize through a configured profile with an API key for the relevant space.
This skill
ax profiles
Saves or replaces Arize profile settings (irreversible)
Arize
Uploads experiment runs to Arize