DeepSeek-V2 is the stronger all-rounder.
Scores updated · 21 tests both models report · How we compare
Where each one wins
Tests won in each of the one area where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Reasoning rests on a single test.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-V2 pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+3.5points ahead
Where DeepSeek-V3.2-Exp-Base pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results20 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- CRUXEval-O (output prediction)DeepSeek-V3.2-Exp-Base by 25.149.874.9+25.1
- Chinese SimpleQA (C-SimpleQA)DeepSeek-V3.2-Exp-Base by 19.548.568+19.5
- MGSMDeepSeek-V3.2-Exp-Base by 18.763.682.3+18.7
- HumanEvalDeepSeek-V3.2-Exp-Base by 18.643.361.9+18.6
- MATHDeepSeek-V3.2-Exp-Base by 16.743.460.1+16.7
- MMLU-ReduxDeepSeek-V3.2-Exp-Base by 12.577.990.4+12.5
- MMLUDeepSeek-V3.2-Exp-Base by 12.275.687.8+12.2
- MMLU-ProDeepSeek-V3.2-Exp-Base by 11.951.463.3+11.9
- CRUXEval-I (input prediction)DeepSeek-V3.2-Exp-Base by 11.452.563.9+11.4
- MBPPDeepSeek-V3.2-Exp-Base by 10.66575.6+10.6
- BBHDeepSeek-V3.2-Exp-Base by 9.978.888.7+9.9
- C-EvalDeepSeek-V3.2-Exp-Base by 9.681.491+9.6
- DROPDeepSeek-V3.2-Exp-Base by 6.580.186.6+6.5
- GSM8KDeepSeek-V3.2-Exp-Base by 5.279.284.4+5.2
- CMMLUDeepSeek-V3.2-Exp-Base by 4.98488.9+4.9
- TriviaQADeepSeek-V3.2-Exp-Base by 479.983.9+4
- ARC-ChallengeDeepSeek-V3.2-Exp-Base by 392.295.2+3
- WinoGrandeDeepSeek-V2 by 2.986.383.4+2.9
- HellaSwagDeepSeek-V3.2-Exp-Base by 2.387.189.4+2.3
- PIQADeepSeek-V3.2-Exp-Base by 1.283.985.1+1.2
Questions people ask
Which is better, DeepSeek-V2 or DeepSeek-V3.2-Exp-Base?
DeepSeek-V2 wins the one area where both have results: reasoning. DeepSeek-V3.2-Exp-Base wins none.
How do you compare the two?
We use the 21 benchmark tests both models have published scores on. The verdict counts the 1 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.