llama-r: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1131 ± 36, rank #10311 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 25.25 | Normalized accuracy (%) | 75.9 |
| Open Korean LLM - KorNAT-Harmless | 61.78 | Normalized accuracy (%) | 34 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 45.38 | MC2 (%) (0-shot) | 30.7 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open Korean LLM - KorNAT-Helpful | 43.45 | Normalized accuracy (%) | 23.6 |
| Open Korean LLM - Ko-IFEval | 18.79 | Strict accuracy, prompt/instruction mean (%) | 14.4 |
| Open LLM Leaderboard v1 - GSM8K | 0.61 | Accuracy (%) (5-shot) | 13.1 |
| Open LLM Leaderboard v1 - MMLU | 26.13 | Accuracy (%) (5-shot) | 11.4 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 0.61 | Exact match, flexible extract (%) | 9.4 |
| Open LLM Leaderboard v1 - WinoGrande | 52.17 | Accuracy (%) (5-shot) | 6.5 |
| Open LLM Leaderboard v1 - HellaSwag | 30.18 | Normalized accuracy (%) (10-shot) | 5.5 |
| Open Korean LLM - Ko-Winogrande | 50.51 | Accuracy (%) | 4 |
Interactive version: theaggregate.ai/model?slug=llama-r · How It Works · Data refreshed daily, snapshot 2026-11-08.