Compare AI models

Specifications
Parameters
3B6B
Context window
128k32k
Benchmarks
HumanEval
77.4%9.8%
MATH
51.7%6.9%
Overview
CompanyMistralZ.ai
Release dateOct 16 2024Jun 25 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Ministral 3B leads ChatGLM2-6B on 2 of the 2 benchmarks they both report (HumanEval, MATH). ChatGLM2-6B shipped 479 days before Ministral 3B, so benchmark comparisons should account for the intervening progress.

Ministral 3B has 3B parameters, while ChatGLM2-6B has 6B. Context windows are 128k (Ministral 3B) vs 32k (ChatGLM2-6B). Ministral 3B is closed, while ChatGLM2-6B is open weight.

On HumanEval, Ministral 3B leads at 77.4% vs ChatGLM2-6B at 9.8%. On MATH, Ministral 3B leads at 51.7% vs ChatGLM2-6B at 6.9%.