Qwen 2.5 14B: benchmark results
Alibaba's open 14B Qwen2.5 mid-size model with a 128K context and improved coding, math, and structured output. Provider: Alibaba. Released 2024-09-19. Access: Open.
Aggregate Rating 1518 ± 9, rank #725 of 1656 rated models, from 386 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open PL LLM - PSC (multiple choice, 5-shot) | 96.48 | Binary F1 (%) | 99.7 |
| Open PL LLM - PPC (generative, 5-shot) | 81.5 | Accuracy (%) | 98.6 |
| Open Arabic LLM - Arabic MMLU Geography (Middle School) | 75 | Accuracy (%) | 98.1 |
| Open Arabic LLM - Arabic MMLU Philosophy (High School) | 71.79 | Accuracy (%) | 97.5 |
| Open PL LLM - PPC (multiple choice, 5-shot) | 81.6 | Accuracy (%) | 96.9 |
| Open Arabic LLM - Arabic MMLU HT Virology | 53.01 | Accuracy (%) | 96.6 |
| Open Arabic LLM - Arabic MMLU Computer Science (Middle School) | 96.3 | Accuracy (%) | 96.3 |
| Open PL LLM - PolEmo2-OUT (multiple choice, 5-shot) | 79.76 | Accuracy (%) | 95.5 |
| Open LLM Leaderboard - GPQA | 38.17 | Score | 95.3 |
| Open Arabic LLM - Arabic MMLU General Knowledge (Primary School) | 77.78 | Accuracy (%) | 93.5 |
| Open Arabic LLM - Arabic MMLU History (High School) | 53.42 | Accuracy (%) | 92.6 |
| Open PL LLM - PoQuAD Open Book (generative, 5-shot) | 71.3 | Levenshtein Similarity (%) | 92.5 |
Interactive version: theaggregate.ai/model?slug=qwen-2-5-14b · How It Works · Data refreshed daily, snapshot 2026-10-20.