Compare AI models

Specifications
Parameters
3B7B
Context window
128k64k
Benchmarks
HumanEval
77.4%83.5%
Overview
CompanyMistralQwen
Release dateOct 16 2024Apr 16 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

CodeQwen1.5 leads Ministral 3B on 1 of the 1 benchmark they both report (HumanEval). CodeQwen1.5 shipped 183 days before Ministral 3B, so benchmark comparisons should account for the intervening progress.

Ministral 3B has 3B parameters, while CodeQwen1.5 has 7B. Context windows are 128k (Ministral 3B) vs 64k (CodeQwen1.5). Ministral 3B is closed, while CodeQwen1.5 is open weight.

On HumanEval, CodeQwen1.5 leads at 83.5% vs Ministral 3B at 77.4%.