DeepSeekMath V2 vs Mercury 2

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

DeepSeekMath V2 finishes one point ahead overall, 66 to 65. That is enough to call, but not enough to treat as a blowout. This matchup comes down to a few meaningful edges rather than one model dominating the board.

DeepSeekMath V2's sharpest advantage is in coding, where it averages 47.3 against 41.1. The single biggest benchmark swing on the page is HLE, 18 to 9. Mercury 2 does hit back in reasoning, so the answer changes if that is the part of the workload you care about most.

Quick Verdict

Pick DeepSeekMath V2 if you want the stronger benchmark profile. Mercury 2 only becomes the better choice if reasoning is the priority.

Agentic

DeepSeekMath V2

DeepSeekMath V2

63.9

Mercury 2

63.7

65
Terminal-Bench 2.0
63
66
BrowseComp
67
61
OSWorld-Verified
62

Coding

DeepSeekMath V2

DeepSeekMath V2

47.3

Mercury 2

41.1

72
HumanEval
75
45
SWE-bench Verified
46
44
LiveCodeBench
38
51
SWE-bench Pro
43

Multimodal & Grounded

Mercury 2

DeepSeekMath V2

68.1

Mercury 2

68.3

64
MMMU-Pro
66
73
OfficeQA Pro
71

Reasoning

Mercury 2

DeepSeekMath V2

75.9

Mercury 2

80.1

77
SimpleQA
82
75
MuSR
82
86
BBH
87
75
LongBench v2
77
72
MRCRv2
76

Knowledge

DeepSeekMath V2

DeepSeekMath V2

61

Mercury 2

57.2

80
MMLU
78
79
GPQA
78
77
SuperGPQA
76
75
OpenBookQA
74
74
MMLU-Pro
72
18
HLE
9
73
FrontierScience
69

Instruction Following

Mercury 2

DeepSeekMath V2

83

Mercury 2

84

83
IFEval
84

Multilingual

DeepSeekMath V2

DeepSeekMath V2

82.5

Mercury 2

79.7

87
MGSM
81
80
MMLU-ProX
79

Mathematics

DeepSeekMath V2

DeepSeekMath V2

84

Mercury 2

80.9

80
AIME 2023
81
82
AIME 2024
83
81
AIME 2025
82
76
HMMT Feb 2023
77
78
HMMT Feb 2024
79
77
HMMT Feb 2025
78
79
BRUMO 2025
80
90
MATH-500
82

Frequently Asked Questions

Which is better, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 is ahead overall, 66 to 65. The biggest single separator in this matchup is HLE, where the scores are 18 and 9.

Which is better for knowledge tasks, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 has the edge for knowledge tasks in this comparison, averaging 61 versus 57.2. Inside this category, HLE is the benchmark that creates the most daylight between them.

Which is better for coding, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 has the edge for coding in this comparison, averaging 47.3 versus 41.1. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.

Which is better for math, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 has the edge for math in this comparison, averaging 84 versus 80.9. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.

Which is better for reasoning, DeepSeekMath V2 or Mercury 2?

Mercury 2 has the edge for reasoning in this comparison, averaging 80.1 versus 75.9. Inside this category, MuSR is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 has the edge for agentic tasks in this comparison, averaging 63.9 versus 63.7. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, DeepSeekMath V2 or Mercury 2?

Mercury 2 has the edge for multimodal and grounded tasks in this comparison, averaging 68.3 versus 68.1. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, DeepSeekMath V2 or Mercury 2?

Mercury 2 has the edge for instruction following in this comparison, averaging 84 versus 83. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, DeepSeekMath V2 or Mercury 2?

DeepSeekMath V2 has the edge for multilingual tasks in this comparison, averaging 82.5 versus 79.7. Inside this category, MGSM is the benchmark that creates the most daylight between them.

Last updated: March 12, 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.