Compare AI models

Specifications
Parameters
8B7B
Context window
128k32k
Benchmarks
HumanEval
76.8%38.4%
MATH
54.5%13.2%
Overview
CompanyMistralMistral
Release dateOct 16 2024May 22 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Ministral 8B leads Mistral 7B v0.3 on 2 of the 2 benchmarks they both report (HumanEval, MATH). Mistral 7B v0.3 shipped 147 days before Ministral 8B, so benchmark comparisons should account for the intervening progress.

Ministral 8B has 8B parameters, while Mistral 7B v0.3 has 7B. Context windows are 128k (Ministral 8B) vs 32k (Mistral 7B v0.3).

On HumanEval, Ministral 8B leads at 76.8% vs Mistral 7B v0.3 at 38.4%. On MATH, Ministral 8B leads at 54.5% vs Mistral 7B v0.3 at 13.2%.