Compare AI models

Specifications
Parameters
7B8B
Context window
4k8k
Benchmarks
GSM8K
88.2%58.45%
Overview
CompanyDeepSeekNVIDIA
Release dateFeb 5 2024Aug 21 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-Math leads Mistral NeMo Minitron 8B on 1 of the 1 benchmark they both report (GSM8K). DeepSeek-Math shipped 198 days before Mistral NeMo Minitron 8B, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while Mistral NeMo Minitron 8B has 8B. Context windows are 4k (DeepSeek-Math) vs 8k (Mistral NeMo Minitron 8B).

On GSM8K, DeepSeek-Math leads at 88.2% vs Mistral NeMo Minitron 8B at 58.45%.