Skip to main content

← Dracon Bench leaderboard

Head-to-head

nemotron-3-nano-omni-30b-a3b-reasoning vs nemotron-3.5-bolt-e4b

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 nemotron-3-nano-omni-30b-a3b-reasoning minus nemotron-3.5-bolt-e4b.

Scores

nemotron-3-nano-omni-30b-a3b-reasoning against nemotron-3.5-bolt-e4b.
Suitenemotron-3-nano-omni-30b-a3b-reasoningnemotron-3.5-bolt-e4bDelta
Overall63.8% 6 of 6· #1270.1% 6 of 6· #4-6.3 pts
IFEval-50 prompt90.0% n=10 59.6%–98.2%100.0% n=10 72.2%–100.0%-10.0 pts
IFBench-50 prompt60.0% n=10 31.3%–83.2%40.0% n=10 16.8%–68.7%+20.0 pts
MMLU-Pro-5020.0% n=10 5.7%–51.0%70.0% n=10 39.7%–89.2%-50.0 pts
JSON briefs100.0% n=5 56.6%–100.0%100.0% n=5 56.6%–100.0%tie
GSM8K-5080.0% n=10 49.0%–94.3%70.0% n=10 39.7%–89.2%+10.0 pts
SVG total66/200 blind rubric102/250 blind rubric-7.8 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

18 vs 19

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

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

18/50

nemotron-3.5-bolt-e4b

19/50

clock

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

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

unjudged

no svg

nemotron-3.5-bolt-e4b

26/50
12 1 2 3 6 9 8 7 10 11 DRACON

duo

17 vs 20

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

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

17/50

nemotron-3.5-bolt-e4b

20/50

monalisa

14 vs 16

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

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

14/50

nemotron-3.5-bolt-e4b

16/50

pelican

17 vs 21

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

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

17/50

nemotron-3.5-bolt-e4b

21/50