Compare AI models

Specifications
Parameters
671B1B/4B/12B/27B
Context window
128k—
API pricing
Cheapest input
$0.256DeepInfra$0.04DeepInfra
Cheapest output
$0.712DeepInfra$0.08DeepInfra
$103StreamLake$15.00DeepInfra · bf16
Benchmarks
LiveCodeBench
40.5%29.7%
GPQA Diamond
59.1%42.4%
MMLU
88.5%76.9%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024Mar 12 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Gemma 3 on 3 of the 3 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU). DeepSeek V3 shipped 76 days before Gemma 3, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Gemma 3 has 1B/4B/12B/27B.

On LiveCodeBench, DeepSeek V3 leads at 40.5% vs Gemma 3 at 29.7%. On GPQA Diamond, DeepSeek V3 leads at 59.1% vs Gemma 3 at 42.4%. On MMLU, DeepSeek V3 leads at 88.5% vs Gemma 3 at 76.9%.