Qwen 1.5 1.8B: benchmark results
Alibaba's Qwen1.5-1.8B base model, the beta version of Qwen2, supporting a 32K context window as part of an eight-size family from 0.5B to 72B parameters. Provider: Alibaba. Released 2024-02-04. Access: Open.
Aggregate Rating 1257 ± 16, rank #1531 of 1656 rated models, from 244 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 30.54 | Score | 62.3 |
| Open LLM Leaderboard v1 - GSM8K | 33.59 | Accuracy (%) (5-shot) | 49.6 |
| EuroEval Norwegian Knowledge - Idioms NO | 2.33 | MCC (x100) | 44.5 |
| Open Japanese LLM - Mbpp Pylint Check | 12.85 | Score (%) | 41 |
| Open Japanese LLM - CG | 1.41 | Score (%) | 36.8 |
| MERA - ruHumanEval | 4.76 | pass@1 (%) | 36.4 |
| MERA - ruCodeEval | 2.26 | pass@1 (%) | 35 |
| MERA - LCS | 10.8 | Accuracy (%) | 31.9 |
| Open Arabic LLM - Arabic MMLU HT Machine Learning | 31.25 | Accuracy (%) | 27.2 |
| EuroEval Norwegian Knowledge | 4.43 | Knowledge Average Score (%) | 26.3 |
| Open LLM Leaderboard v1 - MMLU | 46.71 | Accuracy (%) (5-shot) | 25.3 |
| Open Portuguese LLM - OAB Exams | 28.79 | Accuracy (%) | 24.2 |
Interactive version: theaggregate.ai/model?slug=qwen-1-5-1-8b · How It Works · Data refreshed daily, snapshot 2026-10-20.