Compare AI models

Specifications
Parameters
7B6B
Context window
4k32k
Benchmarks
MATH
51.7%6.9%
GSM8K
88.2%25.9%
Overview
CompanyDeepSeekZ.ai
Release dateFeb 5 2024Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-Math leads ChatGLM2-6B on 2 of the 2 benchmarks they both report (MATH, GSM8K). ChatGLM2-6B shipped 225 days before DeepSeek-Math, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while ChatGLM2-6B has 6B. Context windows are 4k (DeepSeek-Math) vs 32k (ChatGLM2-6B).

On MATH, DeepSeek-Math leads at 51.7% vs ChatGLM2-6B at 6.9%. On GSM8K, DeepSeek-Math leads at 88.2% vs ChatGLM2-6B at 25.9%.