Qwen 1.5 1.8B — benchmark results

Provider: Alibaba. Released 2024-02-05. Access: Open.

Unified ELO 1282 ± 12, rank #1573 of 1776 rated models, from 203 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - GPQA7.38Score62.2
Open Japanese LLM - Mbpp Pylint Check12.85Score (%)41
Open Japanese LLM - CG1.41Score (%)36.8
Open Arabic LLM - Arabic MMLU HT Machine Learning31.25Accuracy (%)27.2
EuroEval Norwegian Knowledge4.43Knowledge Average Score (%)26.3
Open Arabic LLM - Arabic MMLU HT College Physics22.55Accuracy (%)24.1
Open Arabic LLM - Arabic MMLU Physics (High School)29.41Accuracy (%)23.8
EuroEval Portuguese NLU - MultiWikiQA PT30.19Reading comprehension Score (%)21
Open LLM Leaderboard - MMLU-Pro9.8Score20.5
Open LLM Leaderboard - BBH9.76Score19.7
Open LLM Leaderboard - MATH Level 53.17Score19.7
EuroEval English Knowledge19.09Knowledge Average Score (%)19.6

Interactive version: theaggregate.ai/model?slug=qwen-1-5-1-8b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.