Sansa Bench
Engineering leaderboard for Sansa Bench, with charts, model comparison, and methodology. This dimension covers design principles and problem solving across mechanical, electrical, civil, and related fields.
Top models for Engineering
- 1.
Gemini-3-Flash-Preview Reasoning High1.000 - 2.Gemini-3-Pro-Preview Reasoning High1.000
- 3.
Glm-5.2 Reasoning High0.904 - 4.
Qwen3.6-Plus Reasoning Base0.896 - 5.
Gpt-5.2 Reasoning High0.894
Methodology: Engineering
What it measures
Tests engineering knowledge and understanding, including various engineering disciplines, design principles, problem-solving approaches, and engineering applications. Queries cover mechanical, electrical, civil, and other engineering domains.
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.