Peagle-9B: benchmark results
maxcurrent's Peagle 9B, a VortexMerge passthrough splice of mlabonne's NeuralBeagle14-7B and eldogbbhed's NeuralPearlBeagle layers. Provider: maxcurrent. Released 2024-04-01. Access: Open.
Aggregate Rating 1435 ± 15, rank #1054 of 1635 rated models, from 39 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 73.75 | Normalized accuracy (%) | 96.7 |
| Open Korean LLM - KorNAT-Helpful | 52.48 | Normalized accuracy (%) | 93.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 70.16 | MC2 (%) (0-shot) | 90 |
| Open LLM Leaderboard v1 - ARC Challenge | 71.5 | Normalized accuracy (%) (25-shot) | 89.4 |
| Open LLM Leaderboard v1 - HellaSwag | 87.34 | Normalized accuracy (%) (10-shot) | 85.7 |
| Open Chinese LLM - TruthfulQA MC | 59.78 | Accuracy (%) | 82 |
| Open Korean LLM Leaderboard | 41.22 | Average Score (%) | 81.8 |
| Open LLM Leaderboard v1 - GSM8K | 66.11 | Accuracy (%) (5-shot) | 81.2 |
| Open Korean LLM - KorNAT-CKA | 29.34 | Normalized accuracy (%) | 79.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 41.47 | Exact match, flexible extract (%) | 76.1 |
| Open LLM Leaderboard v1 - WinoGrande | 80.35 | Accuracy (%) (5-shot) | 73.5 |
| Open Korean LLM - Ko-EQ-Bench | 25.4 | EQ-Bench score | 71.3 |
Interactive version: theaggregate.ai/model?slug=peagle-9b · How It Works · Data refreshed daily, snapshot 2026-11-05.