DeepSeek LLM 2.0 vs DeepSeek V3.1 (Reasoning)

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

Quick Verdict

DeepSeek LLM 2.0 wins overall with a score of 63 vs 25 (38 point difference).DeepSeek LLM 2.0 wins 4 out of 4 categories.

Knowledge

DeepSeek LLM 2.0

DeepSeek LLM 2.0

76.8

DeepSeek V3.1 (Reasoning)

31.8

79
MMLU
34
78
GPQA
33
76
SuperGPQA
31
74
OpenBookQA
29

Coding

DeepSeek LLM 2.0

DeepSeek LLM 2.0

73

DeepSeek V3.1 (Reasoning)

26

73
HumanEval
26

Mathematics

DeepSeek LLM 2.0

DeepSeek LLM 2.0

79

DeepSeek V3.1 (Reasoning)

33

80
AIME 2023
34
82
AIME 2024
36
81
AIME 2025
35
76
HMMT Feb 2023
30
78
HMMT Feb 2024
32
77
HMMT Feb 2025
31
79
BRUMO 2025
33

Reasoning

DeepSeek LLM 2.0

DeepSeek LLM 2.0

76

DeepSeek V3.1 (Reasoning)

31

77
SimpleQA
32
75
MuSR
30

Frequently Asked Questions

Which is better, DeepSeek LLM 2.0 or DeepSeek V3.1 (Reasoning)?

DeepSeek LLM 2.0 scores higher overall with 63 vs 25, a difference of 38 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek LLM 2.0 or DeepSeek V3.1 (Reasoning)?

DeepSeek LLM 2.0 leads in knowledge tasks with an average score of 76.8 vs 31.8.

Which is better for coding, DeepSeek LLM 2.0 or DeepSeek V3.1 (Reasoning)?

DeepSeek LLM 2.0 leads in coding with an average score of 73 vs 26.

Which is better for math, DeepSeek LLM 2.0 or DeepSeek V3.1 (Reasoning)?

DeepSeek LLM 2.0 leads in math with an average score of 79 vs 33.

Which is better for reasoning, DeepSeek LLM 2.0 or DeepSeek V3.1 (Reasoning)?

DeepSeek LLM 2.0 leads in reasoning with an average score of 76 vs 31.