multimodal benchmark
WorldVQA is a benchmark designed to evaluate atomic vision-centric world knowledge. It assesses models' ability to understand and reason about visual elements representing real-world knowledge.
Updated Aug 7, 2026
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 61.1% | 100.0% | 5 | C | |
| 2 | BY | 53.0% | 75.0% | 5 | C | |
| 3 | MA | 51.0% | 50.0% | 5 | C | |
| 4 | BY | 48.6% | 25.0% | 5 | C | |
| 5 | MA | 46.3% | 0.0% | 5 | C |
Top published rows on the benchmark's original scale.
The top published results on this benchmark's own scale.
Definition and scoring fields from the benchmark registry.
WorldVQA is a benchmark designed to evaluate atomic vision-centric world knowledge. It assesses models' ability to understand and reason about visual elements representing real-world knowledge.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
Common questions about WorldVQA.
Qwen3.7-Plus is currently ranked first with 61.1%.
WorldVQA is a benchmark designed to evaluate atomic vision-centric world knowledge. It assesses models' ability to understand and reason about visual elements representing real-world knowledge.
Yes. Higher values rank better for this benchmark.
5 unique published model results are currently shown.
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.