Claude Sonnet 4 vs DeepSeek-V3.2-Exp

Verdict

DeepSeek-V3.2-Exp leads 3–1 across 4 shared benchmarks.

Claude Sonnet 4 1 · DeepSeek-V3.2-Exp 3 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Claude Sonnet 4DeepSeek-V3.2-Exp

61.3
Aider polyglot
74.2
46.9
Fiction.LiveBench
83.3
33.1
The Agent Company
42.9
46.11
WeirdML
39.48

A higher number is not always a better model. Each score is task performance under a disclosed harness, and some of these benchmarks are saturated or scored under mixed harnesses. Follow any benchmark to its integrity grade before reading a win as decisive.

Follow the record

These models

Claude Sonnet 484.37
DeepSeek-V3.2-Exp83.3