Compare AI models

Specifications
Parameters
16B1.5B/4B
Context window
32k8k
Benchmarks
MMLU
55.7%54.84%
GSM8K
72%54.89%
Overview
CompanyDeepSeekZ.ai
Release dateMay 16 2024Nov 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 Lite leads GLM-Edge on 2 of the 2 benchmarks they both report (MMLU, GSM8K). DeepSeek V2 Lite shipped 197 days before GLM-Edge, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while GLM-Edge has 1.5B/4B. Context windows are 32k (DeepSeek V2 Lite) vs 8k (GLM-Edge).

On MMLU, DeepSeek V2 Lite leads at 55.7% vs GLM-Edge at 54.84%. On GSM8K, DeepSeek V2 Lite leads at 72% vs GLM-Edge at 54.89%.