Alphallama3-8B_v2: benchmark results
Provider: Other. Access: Open.
Unified ELO 1360 ± 17, rank #2171 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GPQA-Diamond | 29.29 | Normalized accuracy (%) | 99.2 |
| Open Korean LLM - KorNAT-CKA | 30.93 | Normalized accuracy (%) | 85.5 |
| Open Korean LLM - KorNAT-Harmless | 63.25 | Normalized accuracy (%) | 59 |
| Open Korean LLM - Ko-EQ-Bench | 1.12 | EQ-Bench score | 46.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 9.02 | Exact match, flexible extract (%) | 32 |
| Open Korean LLM - Ko-Winogrande | 57.22 | Accuracy (%) | 30.4 |
| Open Korean LLM - KorNAT-Helpful | 44.5 | Normalized accuracy (%) | 30.1 |
| Open Korean LLM Leaderboard | 30.51 | Average Score (%) | 19.8 |
| Open Ko-LLM Leaderboard | 30.51 | Average (%) | 19.7 |
| Open Korean LLM - Ko-IFEval | 16.63 | Strict accuracy, prompt/instruction mean (%) | 11.8 |
Interactive version: theaggregate.ai/model?slug=alphallama3-8b-v2 · How It Works · Data refreshed daily, snapshot 2026-09-19.