Mini_Synatra_SFT: benchmark results
Provider: maywell. Access: Open.
Aggregate Rating 1406 ± 28, rank #5788 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-EQ-Bench | 47.92 | EQ-Bench score | 94.6 |
| Open Korean LLM Leaderboard | 42.65 | Average Score (%) | 86.8 |
| Open Korean LLM - KorNAT-Helpful | 49.73 | Normalized accuracy (%) | 81.4 |
| Open Korean LLM - KorNAT-Harmless | 65.5 | Normalized accuracy (%) | 80.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 38.44 | Exact match, flexible extract (%) | 72.9 |
| Open Korean LLM - Ko-Winogrande | 62.98 | Accuracy (%) | 71.6 |
| Open Korean LLM - Ko-IFEval | 32.63 | Strict accuracy, prompt/instruction mean (%) | 63.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 53.67 | MC2 (%) (0-shot) | 60.3 |
| Open LLM Leaderboard v1 - GSM8K | 44.88 | Accuracy (%) (5-shot) | 59 |
| Open LLM Leaderboard v1 - HellaSwag | 83.44 | Normalized accuracy (%) (10-shot) | 58.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.46 | Normalized accuracy (%) (25-shot) | 58 |
| Open LLM Leaderboard v1 - MMLU | 61.2 | Accuracy (%) (5-shot) | 51 |
Interactive version: theaggregate.ai/model?slug=mini-synatra-sft · How It Works · Data refreshed daily, snapshot 2026-11-08.