DeepSeek V3.1 (Reasoning) vs Mistral Large 2

Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.

Quick Verdict

Mistral Large 2 wins overall with a score of 57 vs 25 (32 point difference).Mistral Large 2 wins 4 out of 4 categories.

Knowledge

Mistral Large 2

DeepSeek V3.1 (Reasoning)

31.8

Mistral Large 2

66.5

34
MMLU
68
33
GPQA
68
31
SuperGPQA
66
29
OpenBookQA
64

Coding

Mistral Large 2

DeepSeek V3.1 (Reasoning)

26

Mistral Large 2

60

26
HumanEval
60

Mathematics

Mistral Large 2

DeepSeek V3.1 (Reasoning)

33

Mistral Large 2

67

34
AIME 2023
68
36
AIME 2024
70
35
AIME 2025
69
30
HMMT Feb 2023
64
32
HMMT Feb 2024
66
31
HMMT Feb 2025
65
33
BRUMO 2025
67

Reasoning

Mistral Large 2

DeepSeek V3.1 (Reasoning)

31

Mistral Large 2

65

32
SimpleQA
66
30
MuSR
64

Frequently Asked Questions

Which is better, DeepSeek V3.1 (Reasoning) or Mistral Large 2?

Mistral Large 2 scores higher overall with 57 vs 25, a difference of 32 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek V3.1 (Reasoning) or Mistral Large 2?

Mistral Large 2 leads in knowledge tasks with an average score of 66.5 vs 31.8.

Which is better for coding, DeepSeek V3.1 (Reasoning) or Mistral Large 2?

Mistral Large 2 leads in coding with an average score of 60 vs 26.

Which is better for math, DeepSeek V3.1 (Reasoning) or Mistral Large 2?

Mistral Large 2 leads in math with an average score of 67 vs 33.

Which is better for reasoning, DeepSeek V3.1 (Reasoning) or Mistral Large 2?

Mistral Large 2 leads in reasoning with an average score of 65 vs 31.