Compare AI models

Specifications
Parameters
7B/13B/70B14.3B
Context window
4k32k
Benchmarks
HumanEval
29.9%34.2%
MMLU
68.9%62.5%
GSM8K
56.8%61.5%
Overview
CompanyMetaQwen
Release dateJul 18 2023Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-MoE leads LLaMA 2 on 2 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). LLaMA 2 shipped 254 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

LLaMA 2 has 7B/13B/70B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 4k (LLaMA 2) vs 32k (Qwen1.5-MoE).

On HumanEval, Qwen1.5-MoE leads at 34.2% vs LLaMA 2 at 29.9%. On MMLU, LLaMA 2 leads at 68.9% vs Qwen1.5-MoE at 62.5%. On GSM8K, Qwen1.5-MoE leads at 61.5% vs LLaMA 2 at 56.8%.