Compare AI models

Specifications
Parameters
7B14.3B
Context window
8k32k
Benchmarks
HumanEval
26.2%34.2%
MMLU
62.5%62.5%
GSM8K
36.5%61.5%
Overview
CompanyMistralQwen
Release dateSep 27 2023Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-MoE leads Mistral 7B on 2 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Mistral 7B shipped 183 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Mistral 7B has 7B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 8k (Mistral 7B) vs 32k (Qwen1.5-MoE).

On HumanEval, Qwen1.5-MoE leads at 34.2% vs Mistral 7B at 26.2%. On MMLU, both models score 62.5%. On GSM8K, Qwen1.5-MoE leads at 61.5% vs Mistral 7B at 36.5%.