DeepSeek V4 Flash — Benchmarks
Benchmark scores for DeepSeek V4 Flash aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Overall rank: #29 of 123 open modelscomposite 59.4/100 across 6 benchmarks in 3 categories · methodology
Coding
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SciCode | 44.9% | #17 / 50 | #86 / 155 |
| LiveBench Coding | 69.2 | #15 / 16 | #50 / 54 |
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LiveBench Math | 79.7 | #14 / 16 | #48 / 54 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| LiveBench Reasoning | 70.6 | #15 / 16 | #52 / 54 |
| SimpleBench | 46.3% | #11 / 23 | #60 / 101 |
| CritPt | 7.1% | #11 / 50 | #74 / 167 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.