Head-to-head comparison across 3benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
Hy3 Preview
60
Mellum2-12B-A2.5B-Instruct
27
Pick Hy3 Preview 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.
Coding
+22.8 difference
Knowledge
+5.8 difference
Inst. Following
+12.7 difference
Hy3 Preview
Mellum2-12B-A2.5B-Instruct
$0 / $0
N/A
N/A
N/A
N/A
N/A
256K
128K
Pick Hy3 Preview 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.
Hy3 Preview is clearly ahead on the provisional aggregate, 60 to 27. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Hy3 Preview's sharpest advantage is in coding, where it averages 60 against 37.2. The single biggest benchmark swing on the page is GPQA, 87.2% to 40.9%. 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.
Hy3 Preview 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. Hy3 Preview gives you the larger context window at 256K, compared with 128K for Mellum2-12B-A2.5B-Instruct.
Hy3 Preview is ahead on BenchLM's provisional leaderboard, 60 to 27. The biggest single separator in this matchup is GPQA, where the scores are 87.2% and 40.9%.
Hy3 Preview has the edge for knowledge tasks in this comparison, averaging 46.7 versus 40.9. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Hy3 Preview has the edge for coding in this comparison, averaging 60 versus 37.2. Mellum2-12B-A2.5B-Instruct stays close enough that the answer can still flip depending on your workload.
Mellum2-12B-A2.5B-Instruct has the edge for instruction following in this comparison, averaging 75.8 versus 63.1. Hy3 Preview 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.