Qwen2.5-1M vs LFM2-24B-A2B

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

Qwen2.5-1M is clearly ahead on the aggregate, 66 to 38. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Qwen2.5-1M's sharpest advantage is in reasoning, where it averages 80.9 against 46.6. The single biggest benchmark swing on the page is AIME 2023, 85 to 46.

Qwen2.5-1M gives you the larger context window at 1M, compared with 32K for LFM2-24B-A2B.

Quick Verdict

Pick Qwen2.5-1M 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

Qwen2.5-1M

Qwen2.5-1M

64.7

LFM2-24B-A2B

33.4

65
Terminal-Bench 2.0
30
72
BrowseComp
38
59
OSWorld-Verified
34

Coding

Qwen2.5-1M

Qwen2.5-1M

44.8

LFM2-24B-A2B

18

76
HumanEval
42
47
SWE-bench Verified
18
40
LiveCodeBench
17
49
SWE-bench Pro
19

Multimodal & Grounded

Qwen2.5-1M

Qwen2.5-1M

68.4

LFM2-24B-A2B

41.7

63
MMMU-Pro
39
75
OfficeQA Pro
45

Reasoning

Qwen2.5-1M

Qwen2.5-1M

80.9

LFM2-24B-A2B

46.6

81
SimpleQA
44
79
MuSR
42
82
BBH
63
82
LongBench v2
48
81
MRCRv2
45

Knowledge

Qwen2.5-1M

Qwen2.5-1M

60.4

LFM2-24B-A2B

35.6

84
MMLU
46
83
GPQA
45
81
SuperGPQA
43
79
OpenBookQA
41
74
MMLU-Pro
51
10
HLE
4
74
FrontierScience
43

Instruction Following

Qwen2.5-1M

Qwen2.5-1M

84

LFM2-24B-A2B

68

84
IFEval
68

Multilingual

Qwen2.5-1M

Qwen2.5-1M

80.4

LFM2-24B-A2B

61.4

81
MGSM
64
80
MMLU-ProX
60

Mathematics

Qwen2.5-1M

Qwen2.5-1M

83.6

LFM2-24B-A2B

50.4

85
AIME 2023
46
87
AIME 2024
48
86
AIME 2025
47
81
HMMT Feb 2023
42
83
HMMT Feb 2024
44
82
HMMT Feb 2025
43
84
BRUMO 2025
45
83
MATH-500
57

Frequently Asked Questions

Which is better, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M is ahead overall, 66 to 38. The biggest single separator in this matchup is AIME 2023, where the scores are 85 and 46.

Which is better for knowledge tasks, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for knowledge tasks in this comparison, averaging 60.4 versus 35.6. Inside this category, MMLU is the benchmark that creates the most daylight between them.

Which is better for coding, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for coding in this comparison, averaging 44.8 versus 18. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for math in this comparison, averaging 83.6 versus 50.4. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for reasoning in this comparison, averaging 80.9 versus 46.6. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for agentic tasks in this comparison, averaging 64.7 versus 33.4. 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, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for multimodal and grounded tasks in this comparison, averaging 68.4 versus 41.7. Inside this category, OfficeQA Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for instruction following in this comparison, averaging 84 versus 68. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Qwen2.5-1M or LFM2-24B-A2B?

Qwen2.5-1M has the edge for multilingual tasks in this comparison, averaging 80.4 versus 61.4. 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.