Compare AI models

Specifications
Parameters
7B1.5B/4B
Context window
4k8k
Benchmarks
GSM8K
88.2%54.89%
Overview
CompanyDeepSeekZ.ai
Release dateFeb 5 2024Nov 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-Math leads GLM-Edge on 1 of the 1 benchmark they both report (GSM8K). DeepSeek-Math shipped 298 days before GLM-Edge, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while GLM-Edge has 1.5B/4B. Context windows are 4k (DeepSeek-Math) vs 8k (GLM-Edge).

On GSM8K, DeepSeek-Math leads at 88.2% vs GLM-Edge at 54.89%.