calme-2.2-qwen2-7B: benchmark results
MaziyarPanahi's calme 2.2 fine-tune of Qwen2 7B, trained on OpenHermes-2.5, Orca math word problems, and HelpSteer2 for broad benchmark gains. Provider: MaziyarPanahi. Released 2024-06-27. Access: Open.
Aggregate Rating 1448 ± 21, rank #1016 of 1654 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Portuguese LLM - FaQuAD NLI | 81.98 | Macro F1 (%) | 93.6 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 55.5 | Exact match, flexible extract (%) | 89.3 |
| Open Korean LLM Leaderboard | 42.9 | Average Score (%) | 87.1 |
| Open Korean LLM - Ko-EQ-Bench | 35.91 | EQ-Bench score | 86.1 |
| Open Portuguese LLM - OAB Exams | 52.67 | Accuracy (%) | 82.4 |
| Open Portuguese LLM - BLUEX | 61.2 | Accuracy (%) | 79.9 |
| Open Portuguese LLM - ENEM | 70.19 | Accuracy (%) | 75.2 |
| Open LLM Leaderboard - MuSR | 43.58 | Score | 75 |
| Open LLM Leaderboard - MATH Level 5 | 21.45 | Score | 73.5 |
| Open LLM Leaderboard - MMLU-Pro | 38.99 | Score | 70.5 |
| Open Portuguese LLM - ASSIN2 RTE | 90.89 | Macro F1 (%) | 61.1 |
| Open LLM Leaderboard - BBH | 52.15 | Score | 60 |
Interactive version: theaggregate.ai/model?slug=calme-2-2-qwen2-7b · How It Works · Data refreshed daily, snapshot 2026-10-24.