Compare AI models

Specifications
Parameters
16B8B
Context window
32k8k
Benchmarks
HumanEval
57.3%36.22%
MMLU
55.7%69.51%
GSM8K
72%58.45%
Overview
CompanyDeepSeekNVIDIA
Release dateMay 16 2024Aug 21 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2 Lite leads Mistral NeMo Minitron 8B on 2 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). DeepSeek V2 Lite shipped 97 days before Mistral NeMo Minitron 8B, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Mistral NeMo Minitron 8B has 8B. Context windows are 32k (DeepSeek V2 Lite) vs 8k (Mistral NeMo Minitron 8B).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Mistral NeMo Minitron 8B at 36.22%. On MMLU, Mistral NeMo Minitron 8B leads at 69.51% vs DeepSeek V2 Lite at 55.7%. On GSM8K, DeepSeek V2 Lite leads at 72% vs Mistral NeMo Minitron 8B at 58.45%.