Compare AI models

Specifications
Parameters
16B/236B—
Context window
128k128k
Benchmarks
HumanEval
90.2%75.7%
MATH
75.7%57.9%
MMLU
79.2%81.9%
GSM8K
94.9%90.9%
Overview
CompanyDeepSeekZ.ai
Release dateJun 14 2024Jun 5 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder V2 leads GLM-4-Air on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). GLM-4-Air shipped 9 days before DeepSeek Coder V2, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek Coder V2) vs 128k (GLM-4-Air). DeepSeek Coder V2 is open weight, while GLM-4-Air is closed.

On HumanEval, DeepSeek Coder V2 leads at 90.2% vs GLM-4-Air at 75.7%. On MATH, DeepSeek Coder V2 leads at 75.7% vs GLM-4-Air at 57.9%. On MMLU, GLM-4-Air leads at 81.9% vs DeepSeek Coder V2 at 79.2%. On GSM8K, DeepSeek Coder V2 leads at 94.9% vs GLM-4-Air at 90.9%.