Nemotron-4 15B vs LFM2-24B-A2B

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

Nemotron-4 15B is clearly ahead on the aggregate, 47 to 38. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Nemotron-4 15B's sharpest advantage is in multilingual, where it averages 72.4 against 61.4. The single biggest benchmark swing on the page is MATH-500, 71 to 57.

Quick Verdict

Pick Nemotron-4 15B if you want the stronger benchmark profile. LFM2-24B-A2B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.

Agentic

Nemotron-4 15B

Nemotron-4 15B

41.3

LFM2-24B-A2B

33.4

37
Terminal-Bench 2.0
30
47
BrowseComp
38
42
OSWorld-Verified
34

Coding

Nemotron-4 15B

Nemotron-4 15B

26.5

LFM2-24B-A2B

18

46
HumanEval
42
31
SWE-bench Verified
18
22
LiveCodeBench
17
30
SWE-bench Pro
19

Multimodal & Grounded

Nemotron-4 15B

Nemotron-4 15B

49.6

LFM2-24B-A2B

41.7

46
MMMU-Pro
39
54
OfficeQA Pro
45

Reasoning

Nemotron-4 15B

Nemotron-4 15B

53.4

LFM2-24B-A2B

46.6

52
SimpleQA
44
50
MuSR
42
73
BBH
63
52
LongBench v2
48
51
MRCRv2
45

Knowledge

Nemotron-4 15B

Nemotron-4 15B

42.6

LFM2-24B-A2B

35.6

54
MMLU
46
53
GPQA
45
51
SuperGPQA
43
49
OpenBookQA
41
63
MMLU-Pro
51
5
HLE
4
50
FrontierScience
43

Instruction Following

Nemotron-4 15B

Nemotron-4 15B

79

LFM2-24B-A2B

68

79
IFEval
68

Multilingual

Nemotron-4 15B

Nemotron-4 15B

72.4

LFM2-24B-A2B

61.4

75
MGSM
64
71
MMLU-ProX
60

Mathematics

Nemotron-4 15B

Nemotron-4 15B

61.1

LFM2-24B-A2B

50.4

54
AIME 2023
46
56
AIME 2024
48
55
AIME 2025
47
50
HMMT Feb 2023
42
52
HMMT Feb 2024
44
51
HMMT Feb 2025
43
53
BRUMO 2025
45
71
MATH-500
57

Frequently Asked Questions

Which is better, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B is ahead overall, 47 to 38. The biggest single separator in this matchup is MATH-500, where the scores are 71 and 57.

Which is better for knowledge tasks, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for knowledge tasks in this comparison, averaging 42.6 versus 35.6. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.

Which is better for coding, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for coding in this comparison, averaging 26.5 versus 18. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for math, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for math in this comparison, averaging 61.1 versus 50.4. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for reasoning in this comparison, averaging 53.4 versus 46.6. Inside this category, BBH is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for agentic tasks in this comparison, averaging 41.3 versus 33.4. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for multimodal and grounded tasks in this comparison, averaging 49.6 versus 41.7. Inside this category, OfficeQA Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for instruction following in this comparison, averaging 79 versus 68. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Nemotron-4 15B or LFM2-24B-A2B?

Nemotron-4 15B has the edge for multilingual tasks in this comparison, averaging 72.4 versus 61.4. 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.