Skip to content

Benchmark

RULER · 1M

Long context benchmarks measure how well models use information in long inputs. Scores are percentages. Higher is better.

Models measured
2
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

92.5%
Mean
91.8%
Median
1.3%
Std dev
91.6%
10th percentile
94.7%
90th percentile
91.6%
Min
91.6%94.7%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Nemotron 3 UltraNeMo Evaluator SDK · 1M· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.94.7%
2Nemotron 3 Super 120B A12Bofficial blog Figure 3 · rounded· Reported by NVIDIA Developer· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.91.8%
3Nemotron 3 Super 120B A12Bofficial HF model card · dedicated RULER container· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.91.8%
4Nemotron 3 Super 120B A12Btechnical report Table 5/8 · dedicated RULER container· Reported by NVIDIA· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.91.6%

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.