Compare AI models

Specifications
Parameters
3B14.3B
Context window
128k32k
Benchmarks
HumanEval
77.4%34.2%
Overview
CompanyMistralQwen
Release dateOct 16 2024Mar 28 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Ministral 3B leads Qwen1.5-MoE on 1 of the 1 benchmark they both report (HumanEval). Qwen1.5-MoE shipped 202 days before Ministral 3B, so benchmark comparisons should account for the intervening progress.

Ministral 3B has 3B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 128k (Ministral 3B) vs 32k (Qwen1.5-MoE). Ministral 3B is closed, while Qwen1.5-MoE is open weight.

On HumanEval, Ministral 3B leads at 77.4% vs Qwen1.5-MoE at 34.2%.