WebSquareAI-Instruct-llama-3-8B-v0.5.37: benchmark results
Provider: Other. Access: Open.
Unified ELO 1504 ± 17, rank #1072 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM Leaderboard | 42.25 | Average Score (%) | 85.5 |
| Open Ko-LLM Leaderboard | 42.25 | Average (%) | 85.4 |
| Open Korean LLM - Ko-EQ-Bench | 34.55 | EQ-Bench score | 84.4 |
| Open Korean LLM - Ko-IFEval | 38.91 | Strict accuracy, prompt/instruction mean (%) | 80.5 |
| Open Korean LLM - KorNAT-Helpful | 49.65 | Normalized accuracy (%) | 80.5 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 41.32 | Exact match, flexible extract (%) | 75.9 |
| Open Korean LLM - Ko-GPQA-Diamond | 24.24 | Normalized accuracy (%) | 62.4 |
| Open Korean LLM - Ko-Winogrande | 60.22 | Accuracy (%) | 52.3 |
| Open Korean LLM - KorNAT-CKA | 24.25 | Normalized accuracy (%) | 50.5 |
| Open Korean LLM - KorNAT-Harmless | 61.82 | Normalized accuracy (%) | 34.5 |
Interactive version: theaggregate.ai/model?slug=websquareai-instruct-llama-3-8b-v0-5-37 · How It Works · Data refreshed daily, snapshot 2026-09-19.