TestCraftEvals

Opening your workspace

Evaluation library

Actions

Your first evaluation starts here.

Add the HTML, define what a good output looks like, and choose how to measure it.

0 evaluations

Review history

    Request inclusion

    OpenAI API key

    Used by TestCraft for generation and OpenAI for grading. Kept only in this tab and active runs. Enter it again after refreshing.

    New evaluation

    1. 1Parameters & input
    2. 2Evaluation criteria

    Parameters

    Create Run

    Generation models

    Eval IDEvaluation nameSuiteEndpointMethod

    Clear workspace?

      Permanently remove all suites, evaluations, and run history from the selected workspace. This cannot be undone. You can load the workshop templates again.

      Delete evaluation?

      Delete this run?

      This permanently removes this run’s outputs, scores, and evidence. Your evaluations stay saved. Used model calls are not restored. This cannot be undone.