SOLAR-10.7B-orca-alpaca-gpt4-math: benchmark results
Provider: Upstage. Access: Open.
Unified ELO 1479 ± 17, rank #1299 of 2656 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-Winogrande | 67.17 | Accuracy (%) | 90.6 |
| Open Korean LLM Leaderboard | 40.27 | Average Score (%) | 76 |
| Open Ko-LLM Leaderboard | 40.27 | Average (%) | 75.9 |
| Open Korean LLM - KorNAT-CKA | 27.65 | Normalized accuracy (%) | 71 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 36.16 | Exact match, flexible extract (%) | 69.6 |
| Open Korean LLM - Ko-IFEval | 33.34 | Strict accuracy, prompt/instruction mean (%) | 66 |
| Open Korean LLM - Ko-EQ-Bench | 18.79 | EQ-Bench score | 65 |
| Open Korean LLM - KorNAT-Helpful | 48.02 | Normalized accuracy (%) | 63.8 |
| Open Korean LLM - KorNAT-Harmless | 63.23 | Normalized accuracy (%) | 58.6 |
| Open Korean LLM - Ko-GPQA-Diamond | 21.21 | Normalized accuracy (%) | 24 |
Interactive version: theaggregate.ai/model?slug=solar-10-7b-orca-alpaca-gpt4-math · How It Works · Data refreshed daily, snapshot 2026-09-19.