multimodal benchmark
PerceptionTest
A novel multimodal video benchmark designed to evaluate perception and reasoning skills of pre-trained models across video, audio, and text modalities. Contains 11.6k real-world videos (average 23 seconds) filmed by participants worldwide, densely annotated with six types of labels. Focuses on skills (Memory, Abstraction, Physics, Semantics) and reasoning types (descriptive, explanatory, predictive, counterfactual). Shows significant performance gap between human baseline (91.4%) and state-of-the-art video QA models (46.2%).
Updated Aug 11, 2026
Models2
Model coverage2
MetricScore
EvidenceB
PerceptionTest Ranking
Higher score ranks better on this benchmark.
| 1 | AC | 73.2% | 100.0% | 2 | C | |
| 2 | AC | 70.5% | 0.0% | 2 | C |
PerceptionTest Score Distribution
A closer view of the leading scores on this benchmark.
PerceptionTest Highlights
The leading models and scores on this benchmark.
What is PerceptionTest?
What PerceptionTest measures and how its scores work.
A novel multimodal video benchmark designed to evaluate perception and reasoning skills of pre-trained models across video, audio, and text modalities. Contains 11.6k real-world videos (average 23 seconds) filmed by participants worldwide, densely annotated with six types of labels. Focuses on skills (Memory, Abstraction, Physics, Semantics) and reasoning types (descriptive, explanatory, predictive, counterfactual). Shows significant performance gap between human baseline (91.4%) and state-of-the-art video QA models (46.2%).
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- PerceptionTest
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- perceptiontest|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about PerceptionTest.
Which model scores highest on PerceptionTest?
Qwen2.5 VL 72B Instruct is currently ranked first with 73.2%.
What does PerceptionTest measure?
A novel multimodal video benchmark designed to evaluate perception and reasoning skills of pre-trained models across video, audio, and text modalities. Contains 11.6k real-world videos (average 23 seconds) filmed by participants worldwide, densely annotated with six types of labels. Focuses on skills (Memory, Abstraction, Physics, Semantics) and reasoning types (descriptive, explanatory, predictive, counterfactual). Shows significant performance gap between human baseline (91.4%) and state-of-the-art video QA models (46.2%).
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.