Qwen 1.5 14B Chat: benchmark results
Chat-tuned Qwen1.5 14B checkpoint. Provider: Alibaba. Released 2024-02-04. Access: Open.
Aggregate Rating 1454 ± 9, rank #977 of 1656 rated models, from 379 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Helpful | 52.73 | Normalized accuracy (%) | 93.7 |
| Open Korean LLM - KorNAT-Harmless | 70.06 | Normalized accuracy (%) | 93.3 |
| Open Korean LLM - Ko-EQ-Bench | 42.87 | EQ-Bench score | 92.6 |
| OpenCompass Reasoning - Inductive - Chinese (CompassBench 2405) | 37.4 | Score (%) | 92.1 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 75.98 | Character F1 (%) | 88.5 |
| OpenCompass Agent - CIBench - Chinese (CompassBench 2402) | 57.5 | Score (%) | 88.2 |
| OpenCompass Agent - CIBench - English (CompassBench 2402) | 64.3 | Score (%) | 88.2 |
| OpenCompass LLM - Agent - English (CompassBench 2402) | 66 | Score (%) | 88.2 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 53.9 | Exact match, flexible extract (%) | 87.7 |
| Open Portuguese LLM - ASSIN2 RTE | 93.22 | Macro F1 (%) | 87.6 |
| Open Korean LLM Leaderboard | 42.31 | Average Score (%) | 85.8 |
| Open Ko-LLM Leaderboard | 42.31 | Average (%) | 85.7 |
Interactive version: theaggregate.ai/model?slug=qwen-1-5-14b-chat · How It Works · Data refreshed daily, snapshot 2026-10-20.