Compare AI models

Specifications
Parameters
7B14.3B
Context window
32k32k
Benchmarks
HumanEval
38.4%34.2%
Overview
CompanyMistralQwen
Release dateMay 22 2024Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral 7B v0.3 leads Qwen1.5-MoE on 1 of the 1 benchmark they both report (HumanEval). Qwen1.5-MoE shipped 55 days before Mistral 7B v0.3, so benchmark comparisons should account for the intervening progress.

Mistral 7B v0.3 has 7B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 32k (Mistral 7B v0.3) vs 32k (Qwen1.5-MoE).

On HumanEval, Mistral 7B v0.3 leads at 38.4% vs Qwen1.5-MoE at 34.2%.