Head-to-head comparison across 2benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
LFM2.5-230M
17
MAI-Thinking-1
65
Verified leaderboard positions: LFM2.5-230M unranked · MAI-Thinking-1 #26
Pick MAI-Thinking-1 if you want the stronger benchmark profile. LFM2.5-230M only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
Knowledge
+49.6 difference
Inst. Following
+13.3 difference
LFM2.5-230M
MAI-Thinking-1
$0 / $0
N/A
N/A
N/A
N/A
N/A
32K
256K
Pick MAI-Thinking-1 if you want the stronger benchmark profile. LFM2.5-230M only becomes the better choice if you would rather avoid the extra latency and token burn of a reasoning model.
MAI-Thinking-1 is clearly ahead on the provisional aggregate, 65 to 17. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
MAI-Thinking-1's sharpest advantage is in knowledge, where it averages 69.9 against 20.3. The single biggest benchmark swing on the page is MMLU-Pro, 20.3% to 85%.
MAI-Thinking-1 is the reasoning model in the pair, while LFM2.5-230M 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. MAI-Thinking-1 gives you the larger context window at 256K, compared with 32K for LFM2.5-230M.
MAI-Thinking-1 is ahead on BenchLM's provisional leaderboard, 65 to 17. The biggest single separator in this matchup is MMLU-Pro, where the scores are 20.3% and 85%.
MAI-Thinking-1 has the edge for knowledge tasks in this comparison, averaging 69.9 versus 20.3. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
MAI-Thinking-1 has the edge for instruction following in this comparison, averaging 85 versus 71.7. LFM2.5-230M stays close enough that the answer can still flip depending on your workload.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.