Qwen3.5 397B (Reasoning) vs Ministral 3 14B (Reasoning)

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

Qwen3.5 397B (Reasoning) is clearly ahead on the aggregate, 75 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Qwen3.5 397B (Reasoning)'s sharpest advantage is in coding, where it averages 62.3 against 35. The single biggest benchmark swing on the page is SWE-bench Pro, 65 to 36. Ministral 3 14B (Reasoning) does hit back in multimodal & grounded, so the answer changes if that is the part of the workload you care about most.

Quick Verdict

Pick Qwen3.5 397B (Reasoning) if you want the stronger benchmark profile. Ministral 3 14B (Reasoning) only becomes the better choice if multimodal & grounded is the priority.

Agentic

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

74.8

Ministral 3 14B (Reasoning)

58.5

77
Terminal-Bench 2.0
60
78
BrowseComp
61
70
OSWorld-Verified
55

Coding

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

62.3

Ministral 3 14B (Reasoning)

35

83
HumanEval
62
60
SWE-bench Verified
39
60
LiveCodeBench
33
65
SWE-bench Pro
36

Multimodal & Grounded

Ministral 3 14B (Reasoning)

Qwen3.5 397B (Reasoning)

70.8

Ministral 3 14B (Reasoning)

71.5

64
MMMU-Pro
71
79
OfficeQA Pro
72

Reasoning

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

84.2

Ministral 3 14B (Reasoning)

69.2

87
SimpleQA
70
85
MuSR
70
91
BBH
80
80
LongBench v2
66
82
MRCRv2
66

Knowledge

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

70.1

Ministral 3 14B (Reasoning)

52.1

91
MMLU
71
89
GPQA
70
87
SuperGPQA
68
85
OpenBookQA
66
81
MMLU-Pro
69
29
HLE
7
81
FrontierScience
62

Instruction Following

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

89

Ministral 3 14B (Reasoning)

81

89
IFEval
81

Multilingual

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

87.8

Ministral 3 14B (Reasoning)

77.8

91
MGSM
81
86
MMLU-ProX
76

Mathematics

Qwen3.5 397B (Reasoning)

Qwen3.5 397B (Reasoning)

92.5

Ministral 3 14B (Reasoning)

75.2

93
AIME 2023
73
95
AIME 2024
75
94
AIME 2025
74
89
HMMT Feb 2023
69
91
HMMT Feb 2024
71
90
HMMT Feb 2025
70
92
BRUMO 2025
72
93
MATH-500
79

Frequently Asked Questions

Which is better, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) is ahead overall, 75 to 60. The biggest single separator in this matchup is SWE-bench Pro, where the scores are 65 and 36.

Which is better for knowledge tasks, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for knowledge tasks in this comparison, averaging 70.1 versus 52.1. Inside this category, HLE is the benchmark that creates the most daylight between them.

Which is better for coding, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for coding in this comparison, averaging 62.3 versus 35. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.

Which is better for math, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for math in this comparison, averaging 92.5 versus 75.2. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for reasoning in this comparison, averaging 84.2 versus 69.2. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for agentic tasks in this comparison, averaging 74.8 versus 58.5. 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, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Ministral 3 14B (Reasoning) has the edge for multimodal and grounded tasks in this comparison, averaging 71.5 versus 70.8. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for instruction following in this comparison, averaging 89 versus 81. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Qwen3.5 397B (Reasoning) or Ministral 3 14B (Reasoning)?

Qwen3.5 397B (Reasoning) has the edge for multilingual tasks in this comparison, averaging 87.8 versus 77.8. 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.