Compare AI models

Specifications
Parameters
141B—
Context window
64k—
API pricing
Cheapest input
$2.00Mistral—
Cheapest output
$6.00Mistral—
$700Mistral · zdr—
Benchmarks
HumanEval
45.1%72%
MMLU
77.75%81.5%
GSM8K
78.6%87.6%
Overview
CompanyMistralZ.ai
Release dateApr 10 2024Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4 leads Mixtral 8×22B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). GLM-4 shipped 85 days before Mixtral 8×22B, so benchmark comparisons should account for the intervening progress.

Mixtral 8×22B is open weight, while GLM-4 is closed.

On HumanEval, GLM-4 leads at 72% vs Mixtral 8×22B at 45.1%. On MMLU, GLM-4 leads at 81.5% vs Mixtral 8×22B at 77.75%. On GSM8K, GLM-4 leads at 87.6% vs Mixtral 8×22B at 78.6%.