Compare AI models

Specifications
Parameters
671B—
Context window
128k1M
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
SWE-Bench Verified
42%9.6%
MMLU
88.5%78.9%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024May 14 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Gemini 1.5 Flash on 2 of the 2 benchmarks they both report (SWE-Bench Verified, MMLU). Gemini 1.5 Flash shipped 226 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 1M (Gemini 1.5 Flash). DeepSeek V3 is open weight, while Gemini 1.5 Flash is closed.

On SWE-Bench Verified, DeepSeek V3 leads at 42% vs Gemini 1.5 Flash at 9.6%. On MMLU, DeepSeek V3 leads at 88.5% vs Gemini 1.5 Flash at 78.9%.