Llama-3-SauerkrautLM-8B-Instruct: benchmark results
VAGO Solutions' German-English fine-tune of Llama 3 8B Instruct, two-stage DPO-trained on curated German data with Hyperspace.ai. Provider: VAGOsolutions. Released 2024-05-01. Access: Open.
Aggregate Rating 1468 ± 14, rank #895 of 1656 rated models, from 46 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 89.41 | Normalized accuracy (%) (10-shot) | 99.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 73.72 | Normalized accuracy (%) (25-shot) | 98.2 |
| Open Medical LLM - PubMedQA | 78.4 | Accuracy (%) | 97.1 |
| Open Korean LLM - KorNAT-Helpful | 52.93 | Normalized accuracy (%) | 93.9 |
| Open Korean LLM - Ko-IFEval | 46.15 | Strict accuracy, prompt/instruction mean (%) | 90.8 |
| Open Korean LLM - Ko-EQ-Bench | 39.6 | EQ-Bench score | 90.7 |
| Open LLM Leaderboard v1 - MMLU | 68.07 | Accuracy (%) (5-shot) | 90.5 |
| Open LLM Leaderboard - IFEval | 74.45 | Score | 89.5 |
| Open Korean LLM Leaderboard | 42.98 | Average Score (%) | 87.2 |
| Open Ko-LLM Leaderboard | 42.98 | Average (%) | 87.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 66.25 | MC2 (%) (0-shot) | 84.6 |
| Open Medical LLM - MMLU Medical Genetics | 83 | Accuracy (%) | 84.1 |
Interactive version: theaggregate.ai/model?slug=llama-3-sauerkrautlm-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-10-20.