Mercury 2 vs Moonshot v1

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

Mercury 2 is clearly ahead on the aggregate, 65 to 47. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Mercury 2's sharpest advantage is in reasoning, where it averages 80.1 against 55.5. The single biggest benchmark swing on the page is MuSR, 82 to 49.

Mercury 2 is the reasoning model in the pair, while Moonshot v1 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.

Quick Verdict

Pick Mercury 2 if you want the stronger benchmark profile. Moonshot v1 only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.

Agentic

Mercury 2

Mercury 2

63.7

Moonshot v1

42.2

63
Terminal-Bench 2.0
39
67
BrowseComp
49
62
OSWorld-Verified
41

Coding

Mercury 2

Mercury 2

41.1

Moonshot v1

26.4

75
HumanEval
45
46
SWE-bench Verified
34
38
LiveCodeBench
21
43
SWE-bench Pro
30

Multimodal & Grounded

Mercury 2

Mercury 2

68.3

Moonshot v1

52.6

66
MMMU-Pro
49
71
OfficeQA Pro
57

Reasoning

Mercury 2

Mercury 2

80.1

Moonshot v1

55.5

82
SimpleQA
51
82
MuSR
49
87
BBH
73
77
LongBench v2
58
76
MRCRv2
56

Knowledge

Mercury 2

Mercury 2

57.2

Moonshot v1

42.3

78
MMLU
53
78
GPQA
52
76
SuperGPQA
50
74
OpenBookQA
48
72
MMLU-Pro
64
9
HLE
5
69
FrontierScience
49

Instruction Following

Mercury 2

Mercury 2

84

Moonshot v1

77

84
IFEval
77

Multilingual

Mercury 2

Mercury 2

79.7

Moonshot v1

69.8

81
MGSM
73
79
MMLU-ProX
68

Mathematics

Mercury 2

Mercury 2

80.9

Moonshot v1

61

81
AIME 2023
53
83
AIME 2024
55
82
AIME 2025
54
77
HMMT Feb 2023
49
79
HMMT Feb 2024
51
78
HMMT Feb 2025
50
80
BRUMO 2025
52
82
MATH-500
72

Frequently Asked Questions

Which is better, Mercury 2 or Moonshot v1?

Mercury 2 is ahead overall, 65 to 47. The biggest single separator in this matchup is MuSR, where the scores are 82 and 49.

Which is better for knowledge tasks, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for knowledge tasks in this comparison, averaging 57.2 versus 42.3. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for coding in this comparison, averaging 41.1 versus 26.4. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for math in this comparison, averaging 80.9 versus 61. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for reasoning in this comparison, averaging 80.1 versus 55.5. Inside this category, MuSR is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for agentic tasks in this comparison, averaging 63.7 versus 42.2. 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, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for multimodal and grounded tasks in this comparison, averaging 68.3 versus 52.6. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for instruction following in this comparison, averaging 84 versus 77. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Mercury 2 or Moonshot v1?

Mercury 2 has the edge for multilingual tasks in this comparison, averaging 79.7 versus 69.8. Inside this category, MMLU-ProX 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.