DeepSeek-V3 is the stronger all-rounder.
Scores updated · 18 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.
Coding rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-V3 pulls ahead
- Recent programming contest problemsLiveCodeBench v6+12points ahead
Where Hy3-preview-Base pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 18 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results17 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.
- MATHDeepSeek-V3 by 13.990.276.3+13.9
- MMLU-ProDeepSeek-V3 by 10.175.965.8+10.1
- DROPDeepSeek-V3 by 6.191.685.5+6.1
- CRUXEval-I (input prediction)Hy3-preview-Base by 3.967.371.2+3.9
- C-EvalHy3-preview-Base by 3.386.589.8+3.3
- MMLU-ReduxDeepSeek-V3 by 2.289.186.9+2.2
- SuperGPQADeepSeek-V3 by 2.153.751.6+2.1
- Chinese SimpleQA (C-SimpleQA)Hy3-preview-Base by 1.76869.7+1.7
- SimpleQAHy3-preview-Base by 1.624.926.5+1.6
- CRUXEval-O (output prediction)DeepSeek-V3 by 1.469.868.4+1.4
- GSM8KHy3-preview-Base by 1.49495.4+1.4
- MMLUDeepSeek-V3 by 1.188.587.4+1.1
- CMMLUtie88.889.6tie
- MMMLUtie79.480.2tie
- ARC-Challengetie95.396tie
- PIQAtie84.784.4tie
- MBPP+ (EvalPlus-augmented)tie78.878.7tie
Questions people ask
Which is better, DeepSeek-V3 or Hy3-preview-Base?
DeepSeek-V3 wins the one area where both have results: coding. Hy3-preview-Base wins none.
Which is better for coding?
DeepSeek-V3. It wins the one coding test both models report.
How do you compare the two?
We use the 18 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 17 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.