calme-2.3-qwen2-7B: benchmark results
MaziyarPanahi's calme 2.3 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 1457 ± 19, rank #959 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 57.47 | Exact match, flexible extract (%) | 91.6 |
| Open Korean LLM Leaderboard | 43.85 | Average Score (%) | 89.1 |
| Open Ko-LLM Leaderboard | 43.85 | Average (%) | 88.9 |
| Open Korean LLM - Ko-EQ-Bench | 32.67 | EQ-Bench score | 82 |
| Open LLM Leaderboard - MuSR | 44.22 | Score | 79.2 |
| Open Korean LLM - Ko-IFEval | 36.24 | Strict accuracy, prompt/instruction mean (%) | 75.2 |
| Open LLM Leaderboard - MATH Level 5 | 20.69 | Score | 72.1 |
| Open Portuguese LLM - OAB Exams | 48.11 | Accuracy (%) | 68.2 |
| Open Portuguese LLM - ENEM | 67.95 | Accuracy (%) | 68.1 |
| Open Portuguese LLM - FaQuAD NLI | 73.14 | Macro F1 (%) | 67.9 |
| Open Portuguese LLM - BLUEX | 56.19 | Accuracy (%) | 65.5 |
| Open Korean LLM - Ko-Winogrande | 61.09 | Accuracy (%) | 59.7 |
Interactive version: theaggregate.ai/model?slug=calme-2-3-qwen2-7b · How It Works · Data refreshed daily, snapshot 2026-10-23.