DeepSeek V4 Flash 0731 — Benchmarks

Benchmark scores for DeepSeek V4 Flash 0731 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #25 of 123 open modelscomposite 60.9/100 across 11 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
LiveBench Coding75.0#8 / 16#37 / 54
SciCode49.9%#7 / 50#60 / 155

Knowledge

BenchmarkScoreOpen rankAll models
SimpleQA33.6%#12 / 15#49 / 78

Math

BenchmarkScoreOpen rankAll models
LiveBench Math86.8#7 / 16#36 / 54
AIME 2024/202594.4%#6 / 74#39 / 270
ProofBench56.0%#2 / 21#13 / 64

Reasoning

BenchmarkScoreOpen rankAll models
LiveBench Reasoning86.6#3 / 16#24 / 54
GPQA Diamond91.0%#4 / 83#26 / 291
ARC-AGI89.0%#3 / 16#51 / 200
SimpleBench61.1%#1 / 23#27 / 101
CritPt16.6%#5 / 50#48 / 167

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