long context benchmark
RULER Leaderboard
RULER v1 is a synthetic long-context benchmark with 13 official tasks spanning retrieval, multi-hop tracing, aggregation, and QA.
Updated Sep 5, 2026
Models4
Model coverage4
MetricScore
EvidenceB
RULER Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelNV | Score94.70% | Percentile100.00% | Participants4 | EvidenceC | Evaluated |
| Rank02 | ModelNV | Score91.75% | Percentile66.67% | Participants4 | EvidenceC | Evaluated |
| Rank03 | ModelMI | Score87.10% | Percentile33.33% | Participants4 | EvidenceC | Evaluated |
| Rank04 | ModelMI | Score84.10% | Percentile0.00% | Participants4 | EvidenceC | Evaluated |
RULER Highlights
The leading models and scores on this benchmark.
RULER Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for RULER
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is RULER?
What RULER measures and how its scores work.
RULER v1 is a synthetic long-context benchmark based on the public standalone NVIDIA RULER implementation.
It measures how model quality degrades as input length increases using the Score metric, reported as a ratio, across retrieval, multi-hop tracing, aggregation, and QA tasks.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- RULER
- Modality
- text
- Primary category
- long context
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about RULER.
Which model scores highest on RULER?
Nemotron 3 Ultra (550B A55B) is currently ranked first with 94.70%.
What are the top three models on RULER?
The current leaders are Nemotron 3 Ultra (550B A55B) (94.70%), Nemotron 3 Super (120B A12B) (91.75%), and Phi-3.5-MoE-instruct (87.10%).
Which RULER model has the lowest official input price?
Nemotron 3 Super (120B A12B) has the lowest matched official input price at $0.20 input / $0.80 output per 1M tokens.
Which models are fastest among RULER results?
The fastest matched records are Phi-3.5-mini-instruct (23.00 tok/s via Azure).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.