DeepSeek V3.2 (Thinking) vs Nemotron Ultra 253B

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

Quick Verdict

DeepSeek V3.2 (Thinking) wins overall with a score of 69 vs 40 (29 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.

Knowledge

DeepSeek V3.2 (Thinking)

DeepSeek V3.2 (Thinking)

84

Nemotron Ultra 253B

46.8

87
MMLU
49
85
GPQA
48
83
SuperGPQA
46
81
OpenBookQA
44

Coding

DeepSeek V3.2 (Thinking)

DeepSeek V3.2 (Thinking)

79

Nemotron Ultra 253B

41

79
HumanEval
41

Mathematics

DeepSeek V3.2 (Thinking)

DeepSeek V3.2 (Thinking)

86

Nemotron Ultra 253B

48

87
AIME 2023
49
89
AIME 2024
51
88
AIME 2025
50
83
HMMT Feb 2023
45
85
HMMT Feb 2024
47
84
HMMT Feb 2025
46
86
BRUMO 2025
48

Reasoning

DeepSeek V3.2 (Thinking)

DeepSeek V3.2 (Thinking)

82

Nemotron Ultra 253B

46

83
SimpleQA
47
81
MuSR
45

Frequently Asked Questions

Which is better, DeepSeek V3.2 (Thinking) or Nemotron Ultra 253B?

DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 40, a difference of 29 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek V3.2 (Thinking) or Nemotron Ultra 253B?

DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 46.8.

Which is better for coding, DeepSeek V3.2 (Thinking) or Nemotron Ultra 253B?

DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 41.

Which is better for math, DeepSeek V3.2 (Thinking) or Nemotron Ultra 253B?

DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 48.

Which is better for reasoning, DeepSeek V3.2 (Thinking) or Nemotron Ultra 253B?

DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 46.