Claude 3 Haiku vs LFM2.5-1.2B-Thinking

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

Claude 3 Haiku is clearly ahead on the aggregate, 49 to 33. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude 3 Haiku's sharpest advantage is in multimodal & grounded, where it averages 68.7 against 32.4. The single biggest benchmark swing on the page is MMMU-Pro, 70 to 27.

LFM2.5-1.2B-Thinking is the reasoning model in the pair, while Claude 3 Haiku 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. Claude 3 Haiku gives you the larger context window at 200K, compared with 32K for LFM2.5-1.2B-Thinking.

Quick Verdict

Pick Claude 3 Haiku if you want the stronger benchmark profile. LFM2.5-1.2B-Thinking only becomes the better choice if you want the stronger reasoning-first profile.

Agentic

Claude 3 Haiku

Claude 3 Haiku

44

LFM2.5-1.2B-Thinking

34.1

40
Terminal-Bench 2.0
34
53
BrowseComp
37
42
OSWorld-Verified
32

Coding

Claude 3 Haiku

Claude 3 Haiku

19.3

LFM2.5-1.2B-Thinking

8.2

48
HumanEval
17
17
SWE-bench Verified
10
20
LiveCodeBench
9
19
SWE-bench Pro
7

Multimodal & Grounded

Claude 3 Haiku

Claude 3 Haiku

68.7

LFM2.5-1.2B-Thinking

32.4

70
MMMU-Pro
27
67
OfficeQA Pro
39

Reasoning

Claude 3 Haiku

Claude 3 Haiku

59.6

LFM2.5-1.2B-Thinking

38.4

54
SimpleQA
29
52
MuSR
31
74
BBH
67
63
LongBench v2
39
63
MRCRv2
42

Knowledge

Claude 3 Haiku

Claude 3 Haiku

42.8

LFM2.5-1.2B-Thinking

27

56
MMLU
27
56
GPQA
26
54
SuperGPQA
24
52
OpenBookQA
22
63
MMLU-Pro
51
2
HLE
2
50
FrontierScience
31

Instruction Following

Claude 3 Haiku

Claude 3 Haiku

76

LFM2.5-1.2B-Thinking

72

76
IFEval
72

Multilingual

Claude 3 Haiku

Claude 3 Haiku

71.1

LFM2.5-1.2B-Thinking

60.7

73
MGSM
62
70
MMLU-ProX
60

Mathematics

Claude 3 Haiku

Claude 3 Haiku

62.2

LFM2.5-1.2B-Thinking

42.3

56
AIME 2023
28
58
AIME 2024
30
57
AIME 2025
29
52
HMMT Feb 2023
24
54
HMMT Feb 2024
26
53
HMMT Feb 2025
25
55
BRUMO 2025
27
71
MATH-500
61

Frequently Asked Questions

Which is better, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku is ahead overall, 49 to 33. The biggest single separator in this matchup is MMMU-Pro, where the scores are 70 and 27.

Which is better for knowledge tasks, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for knowledge tasks in this comparison, averaging 42.8 versus 27. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for coding in this comparison, averaging 19.3 versus 8.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.

Which is better for math, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for math in this comparison, averaging 62.2 versus 42.3. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.

Which is better for reasoning, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for reasoning in this comparison, averaging 59.6 versus 38.4. Inside this category, SimpleQA is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for agentic tasks in this comparison, averaging 44 versus 34.1. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for multimodal and grounded tasks in this comparison, averaging 68.7 versus 32.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for instruction following in this comparison, averaging 76 versus 72. Inside this category, IFEval is the benchmark that creates the most daylight between them.

Which is better for multilingual tasks, Claude 3 Haiku or LFM2.5-1.2B-Thinking?

Claude 3 Haiku has the edge for multilingual tasks in this comparison, averaging 71.1 versus 60.7. 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.