Claude Sonnet 4vsDeepSeek-V3.2

Claude Sonnet 4
DeepSeek-V3.2
Benchmarks
Nonsense detection
BullshitBench v2
30%Best
13%
Coding
SWE-Bench Verified
72.7%
Science
GPQA Diamond
75.4%
Community preference (code)
Arena Elo (Code)
1361
Overview
CompanyAnthropicDeepSeek
Release dateMay 22 2025Dec 1 2025
AccessProprietaryOpen Weight

Which is better: Claude Sonnet 4 or DeepSeek-V3.2?

Claude Sonnet 4 leads DeepSeek-V3.2 on 1 of the 1 benchmark they both report (BullshitBench v2). Claude Sonnet 4 shipped 193 days before DeepSeek-V3.2, so benchmark comparisons should account for the intervening progress.

Claude Sonnet 4 is proprietary, while DeepSeek-V3.2 is open weight.

On BullshitBench v2, Claude Sonnet 4 leads at 30% vs DeepSeek-V3.2 at 13%.

Frequently asked questions

Claude Sonnet 4 was released by Anthropic on May 22 2025.

DeepSeek-V3.2 was released by DeepSeek on Dec 1 2025.

Claude Sonnet 4 is a proprietary model released by Anthropic. DeepSeek-V3.2 is an open weight model released by DeepSeek.

Other comparisons