calme-2.7-qwen2-7B: benchmark results
MaziyarPanahi's calme-2.7, a fine-tune of Qwen2 7B using the ChatML prompt template. Provider: MaziyarPanahi. Released 2024-06-27. Access: Open.
Aggregate Rating 1409 ± 20, rank #1198 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 62.47 | Exact match, flexible extract (%) | 95.5 |
| Open LLM Leaderboard - MuSR | 48.24 | Score | 95.4 |
| Open Portuguese LLM - ASSIN2 RTE | 93.38 | Macro F1 (%) | 89.1 |
| Open Korean LLM - Ko-IFEval | 35.7 | Strict accuracy, prompt/instruction mean (%) | 73.7 |
| Open Korean LLM Leaderboard | 39.7 | Average Score (%) | 73.5 |
| Open Ko-LLM Leaderboard | 39.7 | Average (%) | 73.3 |
| Open Portuguese LLM - OAB Exams | 49.43 | Accuracy (%) | 72.5 |
| Open Portuguese LLM - ENEM | 68.23 | Accuracy (%) | 69.3 |
| Open Portuguese LLM - BLUEX | 56.88 | Accuracy (%) | 67.6 |
| Open LLM Leaderboard - MMLU-Pro | 37.05 | Score | 60.4 |
| Open LLM Leaderboard - MATH Level 5 | 13.82 | Score | 59.9 |
| Open Korean LLM - KorNAT-Helpful | 47 | Normalized accuracy (%) | 51.7 |
Interactive version: theaggregate.ai/model?slug=calme-2-7-qwen2-7b · How It Works · Data refreshed daily, snapshot 2026-10-23.