Compare AI models

Specifications
Parameters
7B6B
Context window
4k8k
Benchmarks
MATH
51.7%25.7%
GSM8K
88.2%72.3%
Overview
CompanyDeepSeekZ.ai
Release dateFeb 5 2024Oct 27 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-Math leads ChatGLM3-6B on 2 of the 2 benchmarks they both report (MATH, GSM8K). ChatGLM3-6B shipped 101 days before DeepSeek-Math, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while ChatGLM3-6B has 6B. Context windows are 4k (DeepSeek-Math) vs 8k (ChatGLM3-6B).

On MATH, DeepSeek-Math leads at 51.7% vs ChatGLM3-6B at 25.7%. On GSM8K, DeepSeek-Math leads at 88.2% vs ChatGLM3-6B at 72.3%.