Compare AI models

Specifications
Parameters
7B—
Context window
4k128k
Benchmarks
MATH
51.7%57.9%
GSM8K
88.2%90.9%
Overview
CompanyDeepSeekZ.ai
Release dateFeb 5 2024Jun 5 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads DeepSeek-Math on 2 of the 2 benchmarks they both report (MATH, GSM8K). DeepSeek-Math shipped 121 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

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

On MATH, GLM-4-Air leads at 57.9% vs DeepSeek-Math at 51.7%. On GSM8K, GLM-4-Air leads at 90.9% vs DeepSeek-Math at 88.2%.