Compare AI models

Specifications
Parameters
671B1.8B/3.25B
Context window
128k32k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%55.8%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024Dec 6 2023
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Gemini 1.0 Nano on 1 of the 1 benchmark they both report (MMLU). Gemini 1.0 Nano shipped 386 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Gemini 1.0 Nano has 1.8B/3.25B. Context windows are 128k (DeepSeek V3) vs 32k (Gemini 1.0 Nano). DeepSeek V3 is open weight, while Gemini 1.0 Nano is closed.

On MMLU, DeepSeek V3 leads at 88.5% vs Gemini 1.0 Nano at 55.8%.