multimodal benchmark
Minerva
Minerva is a benchmark for complex video reasoning, evaluating models on multi-step reasoning over long and information-dense video content.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
Minerva leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 70.7% | 100.0% | 2 | C | |
| 2 | BY | 65.9% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
Minerva highlights
The top published results on this benchmark's own scale.
What is Minerva?
Definition and scoring fields from the benchmark registry.
Minerva is a benchmark for complex video reasoning, evaluating models on multi-step reasoning over long and information-dense video content.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Minerva
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- minerva|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Minerva.
Which model scores highest on Minerva?
Seed 2.1 Pro is currently ranked first with 70.7%.
What does Minerva measure?
Minerva is a benchmark for complex video reasoning, evaluating models on multi-step reasoning over long and information-dense video content.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.