Sansa Bench
Trivia leaderboard for Sansa Bench, with charts, model comparison, and methodology. This dimension covers broad factual recall across everyday knowledge domains.
Top models for Trivia
- 1.
Claude-Sonnet-4.6 Reasoning Low0.894 - 2.Claude-Opus-4.8 Reasoning High0.888
- 3.Claude-Sonnet-4.6 Reasoning High0.885
- 4.Claude-Opus-5 Reasoning High0.877
- 5.Claude-Sonnet-4.5 Reasoning Low0.779
Methodology: Trivia
What it measures
Tests general trivia and factual knowledge across various domains. Queries evaluate the model's breadth of factual knowledge and ability to recall specific information.
Scoring & Criteria
Returns score 1.0 if the extracted answer exactly matches the expected answer letter (after normalization), otherwise 0.0. The system prompt requests answers in `<answer>X</answer>` format where X is a letter from the provided choices (A, B, C, D, etc.). The grader normalizes for models that include the full choice text instead of just the letter, or that violate the answer tag format from the system prompt. Only one answer is correct.
Evaluation Type: Multiple Choice
Grades multiple choice responses by exact string matching with normalization.