DeepSeek V3 (Nov 2024): benchmark results

Provider: DeepSeek. Released 2024-12-26.

Aggregate Rating 1635 ± 52, rank #678 of 10870 rated models, from 4 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EvalPlus (HumanEval+ & MBPP+)79.8Pass@1 avg (%)96.8
EvalPlus79.8EvalPlus Avg. (imported)87.5
HumanEval+86.6HumanEval+ pass@1 (imported)80.4
MBPP+73MBPP+ pass@1 (imported)62.5

Interactive version: theaggregate.ai/model?slug=deepseek-v3-nov-2024 · How It Works · Data refreshed daily, snapshot 2026-11-12.