Define custom evaluation criteria with percentage weights, ask any question, and get detailed LLM-judged scores.
General Evaluation
Model: Not configured · Judge: Same as active
70%
0%50%100%
API not configured. Go to Settings
Evaluation Criteria
3 criteria · Total: 100%
Accuracy 40%
Completeness 35%
Clarity 25%
1
Weight40%
2
Weight35%
3
Weight25%