Compare AI models

Specifications
Parameters
47B—
Context window
32k—
Benchmarks
HumanEval
40.2%72%
MMLU
70.6%81.5%
GSM8K
58.4%87.6%
Overview
CompanyMistralZ.ai
Release dateDec 8 2023Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4 leads Mixtral 8×7B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Mixtral 8×7B shipped 39 days before GLM-4, so benchmark comparisons should account for the intervening progress.

Mixtral 8×7B is open weight, while GLM-4 is closed.

On HumanEval, GLM-4 leads at 72% vs Mixtral 8×7B at 40.2%. On MMLU, GLM-4 leads at 81.5% vs Mixtral 8×7B at 70.6%. On GSM8K, GLM-4 leads at 87.6% vs Mixtral 8×7B at 58.4%.