Head-to-head comparison across 2benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
Mellum2-12B-A2.5B-Instruct
27
ZAYA1-8B
57
Pick ZAYA1-8B if you want the stronger benchmark profile. Mellum2-12B-A2.5B-Instruct only becomes the better choice if instruction following is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
Knowledge
+32.2 difference
Inst. Following
+1.8 difference
Mellum2-12B-A2.5B-Instruct
ZAYA1-8B
N/A
$0 / $0
N/A
N/A
N/A
N/A
128K
131K
Pick ZAYA1-8B if you want the stronger benchmark profile. Mellum2-12B-A2.5B-Instruct only becomes the better choice if instruction following is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
ZAYA1-8B is clearly ahead on the provisional aggregate, 57 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
ZAYA1-8B's sharpest advantage is in knowledge, where it averages 73.1 against 40.9. The single biggest benchmark swing on the page is GPQA, 40.9% to 71%. Mellum2-12B-A2.5B-Instruct does hit back in instruction following, so the answer changes if that is the part of the workload you care about most.
ZAYA1-8B is the reasoning model in the pair, while Mellum2-12B-A2.5B-Instruct 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. ZAYA1-8B gives you the larger context window at 131K, compared with 128K for Mellum2-12B-A2.5B-Instruct.
ZAYA1-8B is ahead on BenchLM's provisional leaderboard, 57 to 27. The biggest single separator in this matchup is GPQA, where the scores are 40.9% and 71%.
ZAYA1-8B has the edge for knowledge tasks in this comparison, averaging 73.1 versus 40.9. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Mellum2-12B-A2.5B-Instruct has the edge for instruction following in this comparison, averaging 75.8 versus 74. 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.