Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
LFM2.5-1.2B-Thinking is clearly ahead on the aggregate, 33 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
LFM2.5-1.2B-Thinking's sharpest advantage is in agentic, where it averages 34.1 against 22.9. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 34 to 19.
LFM2.5-1.2B-Thinking is the reasoning model in the pair, while Ministral 3 3B 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. Ministral 3 3B gives you the larger context window at 128K, compared with 32K for LFM2.5-1.2B-Thinking.
Pick LFM2.5-1.2B-Thinking if you want the stronger benchmark profile. Ministral 3 3B only becomes the better choice if you need the larger 128K context window or you would rather avoid the extra latency and token burn of a reasoning model.
LFM2.5-1.2B-Thinking
34.1
Ministral 3 3B
22.9
LFM2.5-1.2B-Thinking
8.2
Ministral 3 3B
6.2
LFM2.5-1.2B-Thinking
32.4
Ministral 3 3B
30.4
LFM2.5-1.2B-Thinking
38.4
Ministral 3 3B
30.1
LFM2.5-1.2B-Thinking
27
Ministral 3 3B
24.5
LFM2.5-1.2B-Thinking
72
Ministral 3 3B
67
LFM2.5-1.2B-Thinking
60.7
Ministral 3 3B
59.7
LFM2.5-1.2B-Thinking
42.3
Ministral 3 3B
36
LFM2.5-1.2B-Thinking is ahead overall, 33 to 27. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 34 and 19.
LFM2.5-1.2B-Thinking has the edge for knowledge tasks in this comparison, averaging 27 versus 24.5. Inside this category, MMLU is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for coding in this comparison, averaging 8.2 versus 6.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for math in this comparison, averaging 42.3 versus 36. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for reasoning in this comparison, averaging 38.4 versus 30.1. Inside this category, MuSR is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for agentic tasks in this comparison, averaging 34.1 versus 22.9. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for multimodal and grounded tasks in this comparison, averaging 32.4 versus 30.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for instruction following in this comparison, averaging 72 versus 67. Inside this category, IFEval is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for multilingual tasks in this comparison, averaging 60.7 versus 59.7. Inside this category, MGSM is the benchmark that creates the most daylight between them.
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.