Compare AI models

Specifications
Parameters
7B6B
Context window
32k32k
Benchmarks
HumanEval
38.4%9.8%
MATH
13.2%6.9%
Overview
CompanyMistralZ.ai
Release dateMay 22 2024Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral 7B v0.3 leads ChatGLM2-6B on 2 of the 2 benchmarks they both report (HumanEval, MATH). ChatGLM2-6B shipped 332 days before Mistral 7B v0.3, so benchmark comparisons should account for the intervening progress.

Mistral 7B v0.3 has 7B parameters, while ChatGLM2-6B has 6B. Context windows are 32k (Mistral 7B v0.3) vs 32k (ChatGLM2-6B).

On HumanEval, Mistral 7B v0.3 leads at 38.4% vs ChatGLM2-6B at 9.8%. On MATH, Mistral 7B v0.3 leads at 13.2% vs ChatGLM2-6B at 6.9%.