Compare AI models

Specifications
Parameters
—671B
Context window
200k128k
API pricing
Cheapest input
—$0.256DeepInfra
Cheapest output
—$0.712DeepInfra
—$103StreamLake
Benchmarks
SWE-Bench Verified
22.2%42%
GPQA Diamond
50.4%59.1%
MMLU
88.2%88.5%
Overview
CompanyAnthropicDeepSeek
Release dateMar 4 2024Dec 26 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Claude 3 Opus on 3 of the 3 benchmarks they both report (SWE-Bench Verified, GPQA Diamond, MMLU). Claude 3 Opus shipped 297 days before DeepSeek V3, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude 3 Opus) vs 128k (DeepSeek V3). Claude 3 Opus is closed, while DeepSeek V3 is open weight.

On SWE-Bench Verified, DeepSeek V3 leads at 42% vs Claude 3 Opus at 22.2%. On GPQA Diamond, DeepSeek V3 leads at 59.1% vs Claude 3 Opus at 50.4%. On MMLU, DeepSeek V3 leads at 88.5% vs Claude 3 Opus at 88.2%.