DeepSeek-R1-Zero is the stronger all-rounder.
Scores updated · 17 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas 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 and reasoning rest on a single test each.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-R1-Zero pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1.6points ahead
Where o1-mini pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 17 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results15 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.
- Arena-Hard (GPT-4-1106 judge)o1-mini by 38.49253.6+38.4
- IFEvalo1-mini by 38.284.846.6+38.2
- Codeforceso1-mini by 376 rating points18201444+376 rating
- AlpacaEval2.0 (LC-winrate)o1-mini by 33.157.824.7+33.1
- Chinese SimpleQA (C-SimpleQA)DeepSeek-R1-Zero by 26.140.366.4+26.1
- SimpleQADeepSeek-R1-Zero by 23.3730.3+23.3
- Aider-Polygloto1-mini by 20.732.912.2+20.7
- AIME 2024DeepSeek-R1-Zero by 14.363.677.9+14.3
- MMLU-Proo1-mini by 11.480.368.9+11.4
- MATH-500 (EM)DeepSeek-R1-Zero by 5.99095.9+5.9
- FRAMES (Acc.)DeepSeek-R1-Zero by 5.476.982.3+5.4
- DROP (3-shot F1)DeepSeek-R1-Zero by 5.283.989.1+5.2
- MMLUDeepSeek-R1-Zero by 3.685.288.8+3.6
- CLUEWSCDeepSeek-R1-Zero by 3.289.993.1+3.2
- MMLU-Reduxo1-mini by 1.186.785.6+1.1
Questions people ask
Which is better, o1-mini or DeepSeek-R1-Zero?
DeepSeek-R1-Zero wins all two areas where both have results: coding and reasoning. o1-mini wins none.
Which is better for coding?
DeepSeek-R1-Zero. It wins the one coding test both models report.
How do you compare the two?
We use the 17 benchmark tests both models have published scores on. The verdict counts the 2 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.