Compare AI models

Specifications
Parameters
7B6B
Context window
32k8k
Benchmarks
HumanEval
38.4%35.9%
Overview
CompanyMistralZ.ai
Release dateMay 22 2024Jul 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral 7B v0.3 leads CodeGeeX2-6B on 1 of the 1 benchmark they both report (HumanEval). CodeGeeX2-6B shipped 302 days before Mistral 7B v0.3, so benchmark comparisons should account for the intervening progress.

Mistral 7B v0.3 has 7B parameters, while CodeGeeX2-6B has 6B. Context windows are 32k (Mistral 7B v0.3) vs 8k (CodeGeeX2-6B).

On HumanEval, Mistral 7B v0.3 leads at 38.4% vs CodeGeeX2-6B at 35.9%.