DeepSeek v3 0324 — Benchmarks

Benchmark scores for DeepSeek v3 0324 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #47 of 123 open modelscomposite 52/100 across 9 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
Aider Polyglot55.1%#8 / 18#33 / 69
SciCode35.8%#30 / 50#125 / 155
SWE-bench Verified42.0%#15 / 16#116 / 162

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro81.3%#18 / 119#57 / 259

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202537.8%#44 / 74#190 / 270
MATH Level 575.5%#5 / 32#40 / 108

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond67.6%#31 / 83#168 / 291
SimpleBench27.2%#18 / 23#84 / 101
CritPt0.0%#37 / 50#134 / 167

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.

DeepSeek v3 0324 Benchmarks — Scores & Rankings | llmrun