DeepSeek v2 — Benchmarks

Benchmark scores for DeepSeek v2 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Knowledge

BenchmarkScoreRank
MMLU78.4%#13 / 74
HellaSwag87.1%#4 / 41

Reasoning

BenchmarkScoreRank
BIG-Bench Hard78.8%#5 / 36

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