Skip to content

Benchmark

OmniScience · Accuracy

Scores are percentages. Higher is better.

Models measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

20.8%
Mean
24.1%
Median
3.3%
Std dev
17.5%
10th percentile
24.1%
90th percentile
17.5%
Min
17.5%24.1%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Nemotron 3 UltraNeMo Evaluator SDK · official table· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.24.1%
2Nemotron 3.5 Lightning 30B A3BAA-Omniscience· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.17.5%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.