AEAI Evalquality + grounding lab

today$0.00/ $2

Projects/New project

New project

Define the model, rubric, and dataset for a new evaluation project.

Project ID*?Unique identifier used in URLs and API references. Use kebab-case, e.g. rag-docs-qa.

Unique slug, lowercase, hyphens OK

Name*?Human-readable project name shown in lists and reports.

Description?One-line summary of what AI behavior this project evaluates. Shown in the project card.

Owner*?Team or person responsible for this project's evaluations and rubric maintenance.

Model*?The AI model being evaluated. Select from connected models or add a custom model ID.

Active rubric ID?ID of the rubric currently used for scoring runs in this project. Must match an existing rubric ID exactly.

Total cases?Expected number of test cases in the evaluation dataset. Used for progress tracking and coverage metrics.

AEDashboard Projects Rubrics Eval Runs Datasets Evaluators Regression Human Review Reports Wiki Play Safety Log