Compare AI models

Specifications
Parameters
236B—
Context window
128k—
Benchmarks
HumanEval
81.1%72%
MATH
53.9%47.9%
MMLU
77.8%81.5%
GSM8K
92.2%87.6%
Overview
CompanyDeepSeekZ.ai
Release dateMay 6 2024Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 leads GLM-4 on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). GLM-4 shipped 111 days before DeepSeek V2, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 is open weight, while GLM-4 is closed.

On HumanEval, DeepSeek V2 leads at 81.1% vs GLM-4 at 72%. On MATH, DeepSeek V2 leads at 53.9% vs GLM-4 at 47.9%. On MMLU, GLM-4 leads at 81.5% vs DeepSeek V2 at 77.8%. On GSM8K, DeepSeek V2 leads at 92.2% vs GLM-4 at 87.6%.