Scores updated · 12 tests both models report · How we compare
Every test, side by side
All 12 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results12 tests
- HumanEvalDeepSeek-V3.2-Base by 19.543.362.8+19.5
- MGSM (EM)DeepSeek-V3.2-Base by 17.763.681.3+17.7
- MATH (EM)DeepSeek-V3.2-Base by 17.143.460.5+17.1
- MMLU-ProDeepSeek-V3.2-Base by 14.151.465.5+14.1
- CMath (EM)DeepSeek-V3.2-Base by 13.978.792.6+13.9
- MMLUDeepSeek-V3.2-Base by 12.275.687.8+12.2
- MMLU-ReduxDeepSeek-V3.2-Base by 9.677.987.5+9.6
- GSM8K (EM)DeepSeek-V3.2-Base by 9.581.691.1+9.5
- BBH (EM)DeepSeek-V3.2-Base by 8.878.887.6+8.8
- DROP (F1)DeepSeek-V3.2-Base by 7.880.488.2+7.8
- TriviaQA (EM)DeepSeek-V3.2-Base by 3.38083.3+3.3
- CLUEWSCDeepSeek-V3.2-Base by 1.58283.5+1.5
Questions people ask
How do you compare the two?
We use the 12 benchmark tests both models have published scores on. None of them is in the eight capability areas we count, so this page lists them without a verdict. Each score is the one shown on the model's own page.