diversity_6
- Task Description: Determine if the criteria for diversity jurisdiction have been met based on the citizenships of plaintiffs and defendants and the amounts associated with claims.
- Task Type: Binary classification
- Document Type: fact pattern
- Number of Samples: 306
- Input Length Range: 85-108 tokens
- Evaluation Metrics: accuracy (maximize), balanced_accuracy (maximize), f1_macro (maximize), f1_micro (maximize), valid_predictions_ratio (maximize)
- Tags: civil procedure, rule application, quantitative reasoning
- Paper: LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models
- Dataset Download: https://hazyresearch.stanford.edu/legalbench/
19 submissions
Rank | Model | accuracy | balanced_accuracy | f1_macro | f1_micro | valid_predictions_ratio | Date | Results |
---|---|---|---|---|---|---|---|---|
1 | o3-mini | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-07-09 | View |
2 | grok-4-0709 | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-07-30 | View |
3 | gpt-5-2025-08-07 | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-08-08 | View |
4 | claude-opus-4-1-20250805 | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-08-05 | View |
5 | grok-3-mini | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-07-29 | View |
6 | claude-opus-4-20250514 | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-07-25 | View |
7 | o3 | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 2025-07-08 | View |
8 | o4-mini | 0.993 | 0.993 | 0.993 | 0.993 | 1.000 | 2025-07-08 | View |
9 | openai/gpt-oss-120b | 0.967 | 0.965 | 0.966 | 0.967 | 1.000 | 2025-08-05 | View |
10 | claude-sonnet-4-20250514 | 0.963 | 0.965 | 0.962 | 0.963 | 0.800 | 2025-07-25 | View |
11 | claude-3-5-haiku-20241022 | 0.960 | 0.957 | 0.960 | 0.960 | 1.000 | 2025-08-01 | View |
12 | deepseek-ai/DeepSeek-V3 | 0.910 | 0.905 | 0.908 | 0.910 | 1.000 | 2025-07-10 | View |
13 | meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo | 0.883 | 0.885 | 0.883 | 0.883 | 1.000 | 2025-07-30 | View |
14 | google/gemma-2-27b-it | 0.823 | 0.812 | 0.813 | 0.823 | 1.000 | 2025-07-24 | View |
15 | gpt-4o-mini | 0.803 | 0.802 | 0.802 | 0.803 | 1.000 | 2025-07-02 | View |
16 | deepseek-ai/DeepSeek-R1 | 0.566 | 0.555 | 0.552 | 0.566 | 0.451 | 2025-07-11 | View |
17 | meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo | 0.510 | 0.538 | 0.399 | 0.510 | 1.000 | 2025-07-25 | View |
18 | gpt-4.1-nano | 0.497 | 0.523 | 0.399 | 0.497 | 1.000 | 2025-07-03 | View |
19 | claude-3-haiku-20240307 | 0.487 | 0.516 | 0.354 | 0.487 | 1.000 | 2025-07-25 | View |