Head-to-head comparison across 2benchmark categories. Overall scores shown here use BenchLM's provisional ranking lane.
Hy3 Preview
62
ZAYA1-8B
62
Treat this as a split decision. Hy3 Preview makes more sense if you need the larger 256K context window; ZAYA1-8B is the better fit if knowledge is the priority.
Knowledge
+26.4 difference
Inst. Following
+10.9 difference
Hy3 Preview
ZAYA1-8B
$0 / $0
$0 / $0
N/A
N/A
N/A
N/A
256K
131K
Treat this as a split decision. Hy3 Preview makes more sense if you need the larger 256K context window; ZAYA1-8B is the better fit if knowledge is the priority.
Hy3 Preview and ZAYA1-8B finish on the same provisional overall score, so this is less about a single winner and more about where the edge shows up. The provisional headline says tie; the benchmark table is where the real choice happens.
Hy3 Preview gives you the larger context window at 256K, compared with 131K for ZAYA1-8B.
Hy3 Preview and ZAYA1-8B are tied on the provisional overall score, so the right pick depends on which category matters most for your use case.
ZAYA1-8B has the edge for knowledge tasks in this comparison, averaging 73.1 versus 46.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
ZAYA1-8B has the edge for instruction following in this comparison, averaging 74 versus 63.1. Inside this category, IFBench 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.