multimodal benchmark
Nuscene
A multimodal benchmark for scene understanding and reasoning over the nuScenes autonomous driving domain.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
Nuscene leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 15.4% | 100.0% | 3 | C | |
| 2 | AC | 15.2% | 50.0% | 3 | C | |
| 3 | AC | 14.6% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
Nuscene highlights
The top published results on this benchmark's own scale.
What is Nuscene?
Definition and scoring fields from the benchmark registry.
A multimodal benchmark for scene understanding and reasoning over the nuScenes autonomous driving domain.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Nuscene
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- nuscene|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Nuscene.
Which model scores highest on Nuscene?
Qwen3.5-122B-A10B is currently ranked first with 15.4%.
What does Nuscene measure?
A multimodal benchmark for scene understanding and reasoning over the nuScenes autonomous driving domain.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.