MiMo V2.5 Pro Base is the stronger all-rounder.
Scores updated · 20 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 MiMo V2.5 Pro Base pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+15.7points ahead
- Recent programming contest problemsLiveCodeBench v6+14.8points ahead
Where DeepSeek-V3.1-Base pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 20 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results18 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.
- MATHMiMo V2.5 Pro Base by 23.662.686.2+23.6
- aime_2024_2025_combinedMiMo V2.5 Pro Base by 15.721.637.3+15.7
- HumanEval+MiMo V2.5 Pro Base by 1164.675.6+11
- MMLU-ProMiMo V2.5 Pro Base by 9.758.868.5+9.7
- GSM8KMiMo V2.5 Pro Base by 8.291.499.6+8.2
- MMLU-ReduxMiMo V2.5 Pro Base by 2.89092.8+2.8
- TriviaQADeepSeek-V3.1-Base by 2.283.581.3+2.2
- MMLUMiMo V2.5 Pro Base by 287.489.4+2
- MBPP+MiMo V2.5 Pro Base by 1.972.274.1+1.9
- MBPP+ (EvalPlus-augmented)MiMo V2.5 Pro Base by 1.972.274.1+1.9
- GlobalMMLUMiMo V2.5 Pro Base by 1.781.983.6+1.7
- ARC-ChallengeMiMo V2.5 Pro Base by 1.695.697.2+1.6
- C-EvalMiMo V2.5 Pro Base by 1.59091.5+1.5
- CMMLUMiMo V2.5 Pro Base by 1.488.890.2+1.4
- HellaSwagtie89.289.8tie
- WinoGrandetie85.985.6tie
- BBHtie88.288.4tie
- DROPtie86.386.3tie
Questions people ask
Which is better, DeepSeek-V3.1-Base or MiMo V2.5 Pro Base?
MiMo V2.5 Pro Base wins all two areas where both have results: coding and reasoning. DeepSeek-V3.1-Base wins none.
Which is better for coding?
MiMo V2.5 Pro Base. It wins the one coding test both models report.
How do you compare the two?
We use the 20 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 18 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.