by Orolol · Agent Tool · ★ 44
FamilyBench 🌳 FamilyBench is an evaluation tool for testing the relational reasoning capabilities of Large Language Models (LLMs). It generates random family trees, converts them to textual descriptions, and creates question-answer pairs to evaluate understanding of complex family relationships. 🎯 Objective FamilyBench enables systematic and reproducible evaluation of LLMs' ability to: Understand direct family relationships (parents, children) Infer complex relationships (grandparents, cousins, uncles/aunts) Reason across multiple generations Combine relationships with attributes...
| Stars | 44 |
| Forks | 2 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 58.0124867025186/100 |
| Open Issues | 2 |
| Last Updated | 2026-05-04 |
| Created | 2025-07-29 |
| Platforms | python |
| Est. Tokens | ~512k |
These tools work well together with familyBench for enhanced workflows:
Explore other popular agent tool tools:
familyBench is FamilyBench evaluation tool for testing the relational reasoning capabilities of Large Language Models (LLMs).. It is categorized as a Agent Tool with 44 GitHub stars.
familyBench is primarily written in Python.
You can find installation instructions and usage details in the familyBench GitHub repository at github.com/Orolol/familyBench. The project has 44 stars and 2 forks, indicating an active community.
familyBench is released under the MIT license, making it free to use and modify according to the license terms.