Skip to main content

← Dracon Bench leaderboard

Head-to-head

muse-glimmer-30b vs nemotron-3-nano-omni-30b-a3b-reasoning

The same frozen prompts at temperature 0, scored the same way. Every figure carries the items it was taken over, and the delta is always muse-glimmer-30b minus nemotron-3-nano-omni-30b-a3b-reasoning.

Scores

muse-glimmer-30b against nemotron-3-nano-omni-30b-a3b-reasoning.
Suitemuse-glimmer-30bnemotron-3-nano-omni-30b-a3b-reasoningDelta
Overall47.0% 6 of 6· #2066.1% 6 of 6· #6-19.1 pts
IFEval-50 prompt60.0% n=10 31.3%–83.2%88.0% n=50 76.2%–94.4%-28.0 pts
IFBench-50 prompt20.0% n=10 5.7%–51.0%57.1% n=49 43.3%–70.0%-37.1 pts
MMLU-Pro-5010.0% n=10 1.8%–40.4%32.0% n=50 20.8%–45.8%-22.0 pts
JSON briefs100.0% n=5 56.6%–100.0%100.0% n=5 56.6%–100.0%tie
GSM8K-5010.0% n=10 1.8%–40.4%78.0% n=50 64.8%–87.2%-68.0 pts
SVG total41/50 blind rubric104/250 blind rubric+40.4 pts

Deltas compare scores, not denominators: a 10-point gap means less when one side answered far fewer items. The n= lines keep that honest.

beach

Sun low over water with a reflection path; distant objects smaller and hazier than near ones.

muse-glimmer-30b

unjudged

no svg

nemotron-3-nano-omni-30b-a3b-reasoning

17/50

clock

A clock is checkable to the minute: 10:10 means the hour hand just past 10, the minute hand on 2.

muse-glimmer-30b

unjudged

no svg

nemotron-3-nano-omni-30b-a3b-reasoning

31/50
12 1 2 3 4 5 6 7 8 9 10 11 DRACON

duo

Two characters mid-action, not posed: a dog driving a jet ski, a raccoon holding on.

muse-glimmer-30b

unjudged

no svg

nemotron-3-nano-omni-30b-a3b-reasoning

20/50

monalisa

Leonardo's portrait: three-quarter bust, folded hands, winding river landscape behind.

muse-glimmer-30b

unjudged

no svg

nemotron-3-nano-omni-30b-a3b-reasoning

16/50

pelican

41 vs 20

Everyone knows what a pelican looks like: large beak with throat pouch, long neck, webbed feet.

muse-glimmer-30b

41/50

nemotron-3-nano-omni-30b-a3b-reasoning

20/50