Compare AI models

Specifications
Parameters
236B6B
Context window
128k8k
Benchmarks
HumanEval
84.76%58.5%
MATH
71.02%25.7%
Overview
CompanyDeepSeekZ.ai
Release dateJun 28 2024Oct 27 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2-0628 leads ChatGLM3-6B on 2 of the 2 benchmarks they both report (HumanEval, MATH). ChatGLM3-6B shipped 245 days before DeepSeek V2-0628, so benchmark comparisons should account for the intervening progress.

DeepSeek V2-0628 has 236B parameters, while ChatGLM3-6B has 6B. Context windows are 128k (DeepSeek V2-0628) vs 8k (ChatGLM3-6B).

On HumanEval, DeepSeek V2-0628 leads at 84.76% vs ChatGLM3-6B at 58.5%. On MATH, DeepSeek V2-0628 leads at 71.02% vs ChatGLM3-6B at 25.7%.