Compare AI models

Specifications
Parameters
671B—
Context window
128k32k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
MMLU
88.5%83.7%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024Feb 8 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Gemini 1.0 Ultra on 1 of the 1 benchmark they both report (MMLU). Gemini 1.0 Ultra shipped 322 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 32k (Gemini 1.0 Ultra). DeepSeek V3 is open weight, while Gemini 1.0 Ultra is closed.

On MMLU, DeepSeek V3 leads at 88.5% vs Gemini 1.0 Ultra at 83.7%.