ghost-7B-v0.9.1: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1335 ± 27, rank #8069 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-CKA | 29.41 | Normalized accuracy (%) | 79.4 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Korean LLM - Ko-IFEval | 28.02 | Strict accuracy, prompt/instruction mean (%) | 45.9 |
| Open LLM Leaderboard v1 - GSM8K | 26.91 | Accuracy (%) (5-shot) | 44.9 |
| Open Korean LLM - Ko-EQ-Bench | 0.2 | EQ-Bench score | 43.9 |
| Open LLM Leaderboard v1 - MMLU | 54.78 | Accuracy (%) (5-shot) | 34.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 55.38 | Normalized accuracy (%) (25-shot) | 33.3 |
| Open LLM Leaderboard v1 - HellaSwag | 77.03 | Normalized accuracy (%) (10-shot) | 29.9 |
| Open LLM Leaderboard v1 - WinoGrande | 72.53 | Accuracy (%) (5-shot) | 29.3 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 6.44 | Exact match, flexible extract (%) | 28 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 43.96 | MC2 (%) (0-shot) | 25.4 |
| Open Korean LLM - Ko-Winogrande | 54.54 | Accuracy (%) | 16.6 |
Interactive version: theaggregate.ai/model?slug=ghost-7b-v0-9-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.