Compare AI models

Specifications
Parameters
671B8B
Context window
128k8k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%69.51%
Overview
CompanyDeepSeekNVIDIA
Release dateDec 26 2024Aug 21 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Mistral NeMo Minitron 8B on 1 of the 1 benchmark they both report (MMLU). Mistral NeMo Minitron 8B shipped 127 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Mistral NeMo Minitron 8B has 8B. Context windows are 128k (DeepSeek V3) vs 8k (Mistral NeMo Minitron 8B).

On MMLU, DeepSeek V3 leads at 88.5% vs Mistral NeMo Minitron 8B at 69.51%.