GritLM-7B: benchmark results
Provider: GritLM. Released 2024-02-11. Access: Open.
Aggregate Rating 1382 ± 18, rank #1252 of 1635 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MAPLE - All Aspects | 15.7 | AllAspect@20 (%; share of target papers retrieved in the top | 100 |
| MAPLE - Any Aspect | 98.1 | AnyAspect@20 (%; share of target papers retrieved in the top | 100 |
| MAPLE - Aspect Coverage | 61.8 | AspectCoverage@20 (%; mean share of a paper's aspect queries | 100 |
| MAPLE - Experiment Queries | 54.5 | Recall@20 (%; over MAPLE's 2,095 multi-aspect full-paper sci | 100 |
| MAPLE - Method Queries | 64.6 | Recall@20 (%; over MAPLE's 2,095 multi-aspect full-paper sci | 100 |
| MAPLE - Motivation Queries | 79 | Recall@20 (%; over MAPLE's 2,095 multi-aspect full-paper sci | 100 |
| Open Portuguese LLM - FaQuAD NLI | 80.85 | Macro F1 (%) | 91 |
| Open Korean LLM - KorNAT-Harmless | 64.01 | Normalized accuracy (%) | 69.5 |
| Open Korean LLM - Ko-IFEval | 33.47 | Strict accuracy, prompt/instruction mean (%) | 67.2 |
| Open Portuguese LLM - ASSIN2 RTE | 91.29 | Macro F1 (%) | 65.9 |
| Open Korean LLM - Ko-EQ-Bench | 16.5 | EQ-Bench score | 61.8 |
| Open Korean LLM - KorNAT-Helpful | 47.85 | Normalized accuracy (%) | 61.8 |
Interactive version: theaggregate.ai/model?slug=gritlm-7b · How It Works · Data refreshed daily, snapshot 2026-11-05.