Qwen2.5-14B-Wernicke-SFT: benchmark results
Provider: Alibaba. Released 2024-11-16. Access: Open.
Unified ELO 1591 ± 60, rank #663 of 2656 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Portuguese LLM - FaQuAD NLI | 86.54 | Macro F1 (%) | 99.4 |
| Open Portuguese LLM - ASSIN2 RTE | 94.36 | Macro F1 (%) | 96 |
| Open Portuguese LLM - ENEM | 80.41 | Accuracy (%) | 93.8 |
| Open Portuguese LLM - OAB Exams | 61.32 | Accuracy (%) | 93.4 |
| Open Portuguese LLM - BLUEX | 72.04 | Accuracy (%) | 93.1 |
| Open LLM Leaderboard - BBH | 64.61 | Score | 90.1 |
| Open LLM Leaderboard - GPQA | 35.4 | Score | 89.1 |
| Open LLM Leaderboard - MMLU-Pro | 50.7 | Score | 88.2 |
| Open LLM Leaderboard - MATH Level 5 | 35.95 | Score | 86.3 |
| Open LLM Leaderboard - IFEval | 49.37 | Score | 57.8 |
| Open LLM Leaderboard - MuSR | 39 | Score | 35.2 |
Interactive version: theaggregate.ai/model?slug=qwen2-5-14b-wernicke-sft · How It Works · Data refreshed daily, snapshot 2026-09-19.