Head-to-head comparison across 1benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
LFM2.5-8B-A1B
50
o1
57
Pick o1 if you want the stronger benchmark profile. LFM2.5-8B-A1B only becomes the better choice if you want the cheaper token bill.
Inst. Following
+12.7 difference
LFM2.5-8B-A1B
o1
$0 / $0
$15 / $60
N/A
98 t/s
N/A
32.29s
128K
200K
Pick o1 if you want the stronger benchmark profile. LFM2.5-8B-A1B only becomes the better choice if you want the cheaper token bill.
o1 is clearly ahead on the provisional aggregate, 57 to 50. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
o1's sharpest advantage is in instruction following, where it averages 92.2 against 79.5. The single biggest benchmark swing on the page is IFEval, 91.8% to 92.2%.
o1 is also the more expensive model on tokens at $15.00 input / $60.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for LFM2.5-8B-A1B. That is roughly Infinityx on output cost alone. o1 gives you the larger context window at 200K, compared with 128K for LFM2.5-8B-A1B.
o1 is ahead on BenchLM's provisional leaderboard, 57 to 50. The biggest single separator in this matchup is IFEval, where the scores are 91.8% and 92.2%.
o1 has the edge for instruction following in this comparison, averaging 92.2 versus 79.5. Inside this category, IFEval is the benchmark that creates the most daylight between them.
For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.