Compare AI models

Specifications
Parameters
—671B
Context window
200k128k
API pricing
Cheapest input
—$0.256DeepInfra
Cheapest output
—$0.712DeepInfra
—$103StreamLake
Benchmarks
SWE-Bench Verified
72.5%42%
GPQA Diamond
79.6%59.1%
Overview
CompanyAnthropicDeepSeek
Release dateMay 22 2025Dec 26 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude Opus 4 leads DeepSeek V3 on 2 of the 2 benchmarks they both report (SWE-Bench Verified, GPQA Diamond). DeepSeek V3 shipped 147 days before Claude Opus 4, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude Opus 4) vs 128k (DeepSeek V3). Claude Opus 4 is closed, while DeepSeek V3 is open weight.

On SWE-Bench Verified, Claude Opus 4 leads at 72.5% vs DeepSeek V3 at 42%. On GPQA Diamond, Claude Opus 4 leads at 79.6% vs DeepSeek V3 at 59.1%.