Samantha-Qwen-2-7B: benchmark results
macadeliccc's Samantha-persona conversational fine-tune of Qwen2 7B, QLoRA-trained on Opus Samantha and OpenHermes data. Provider: macadeliccc. Released 2024-06-15. Access: Open.
Aggregate Rating 1424 ± 21, rank #1129 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 47.99 | Score | 94.7 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 59.21 | Exact match, flexible extract (%) | 93.2 |
| Open Korean LLM - KorNAT-Helpful | 50.28 | Normalized accuracy (%) | 84.9 |
| Open Portuguese LLM - FaQuAD NLI | 78.65 | Macro F1 (%) | 84.8 |
| Open Portuguese LLM - ASSIN2 RTE | 92.68 | Macro F1 (%) | 82 |
| Open Korean LLM Leaderboard | 41.15 | Average Score (%) | 81.2 |
| Open Portuguese LLM - BLUEX | 61.61 | Accuracy (%) | 80.7 |
| Open Portuguese LLM - ENEM | 69.77 | Accuracy (%) | 73.9 |
| Open LLM Leaderboard - MATH Level 5 | 21.15 | Score | 72.9 |
| Open Portuguese LLM - OAB Exams | 49.43 | Accuracy (%) | 72.5 |
| Open Korean LLM - Ko-IFEval | 34.37 | Strict accuracy, prompt/instruction mean (%) | 70.8 |
| Open LLM Leaderboard - MMLU-Pro | 37.79 | Score | 64.3 |
Interactive version: theaggregate.ai/model?slug=samantha-qwen-2-7b · How It Works · Data refreshed daily, snapshot 2026-10-24.