solar-ko-recovery-11B-freeze: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1405 ± 17, rank #1869 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 66.77 | Accuracy (%) | 89.5 |
| Open Korean LLM - KorNAT-CKA | 28.48 | Normalized accuracy (%) | 75.8 |
| Open Korean LLM - KorNAT-Helpful | 47.62 | Normalized accuracy (%) | 58.1 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 17.82 | Exact match, flexible extract (%) | 48.7 |
| Open Korean LLM Leaderboard | 33.44 | Average Score (%) | 41.9 |
| Open Ko-LLM Leaderboard | 33.44 | Average (%) | 41.7 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
| Open Korean LLM - KorNAT-Harmless | 61.11 | Normalized accuracy (%) | 23.2 |
| Open Korean LLM - Ko-IFEval | 21.93 | Strict accuracy, prompt/instruction mean (%) | 22.2 |
| Open Korean LLM - Ko-EQ-Bench | -0.97 | EQ-Bench score | 4.1 |
Interactive version: theaggregate.ai/model?slug=solar-ko-recovery-11b-freeze · How It Works · Data refreshed daily, snapshot 2026-09-19.