Skip to main content

← Dracon Bench leaderboard

Head-to-head

lfm-2.5-2.6b 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 lfm-2.5-2.6b minus nemotron-3-nano-omni-30b-a3b-reasoning.

Scores

lfm-2.5-2.6b against nemotron-3-nano-omni-30b-a3b-reasoning.
Suitelfm-2.5-2.6bnemotron-3-nano-omni-30b-a3b-reasoningDelta
Overallunranked63.8% 6 of 6· #12
IFEval-50 prompt86.0% n=50 73.8%–93.0%90.0% n=10 59.6%–98.2%-4.0 pts
IFBench-50 promptn/a60.0% n=10 31.3%–83.2%
MMLU-Pro-50n/a20.0% n=10 5.7%–51.0%
JSON briefsn/a100.0% n=5 56.6%–100.0%
GSM8K-5074.0% n=50 60.4%–84.1%80.0% n=10 49.0%–94.3%-6.0 pts
SVG total68/200 blind rubric66/200 blind rubric+1.0 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

15 vs 18

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

lfm-2.5-2.6b

15/50

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

18/50

clock

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

lfm-2.5-2.6b

unjudged

no svg

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

unjudged

no svg

duo

26 vs 17

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

lfm-2.5-2.6b

26/50

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

17/50

monalisa

11 vs 14

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

lfm-2.5-2.6b

11/50

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

14/50

pelican

16 vs 17

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

lfm-2.5-2.6b

16/50

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

17/50