RAGAs Task

RAGAs-based evaluation feature that allows you to automatically evaluate and quantify the generated responses and retrieved contexts with RAGAs metrics. The overall flow is similar to a general Evaluation Task, but detailed manual editing functions are limited.

---

Create Task

Create a new RAGAs Task.

Run Eval Set

Create and run an Eval Set.

Check Results

Check and analyze RAGAs evaluation results.

BEIR Leaderboard

Check results in the BEIR Leaderboard view.