multimodal benchmark
VQAv2 (val)
VQAv2 is a balanced Visual Question Answering dataset containing open-ended questions about images that require understanding of vision, language, and commonsense knowledge to answer. VQAv2 addresses bias issues from the original VQA dataset by collecting complementary images such that every question is associated with similar images that result in different answers, forcing models to actually understand visual content rather than relying on language priors.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
VQAv2 (val) leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | GO | 71.6% | 100.0% | 3 | C | |
| 2 | GO | 71.0% | 50.0% | 3 | C | |
| 3 | GO | 62.4% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
VQAv2 (val) highlights
The top published results on this benchmark's own scale.
What is VQAv2 (val)?
Definition and scoring fields from the benchmark registry.
VQAv2 is a balanced Visual Question Answering dataset containing open-ended questions about images that require understanding of vision, language, and commonsense knowledge to answer. VQAv2 addresses bias issues from the original VQA dataset by collecting complementary images such that every question is associated with similar images that result in different answers, forcing models to actually understand visual content rather than relying on language priors.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- VQAv2 (val)
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- vqav2-(val)|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about VQAv2 (val).
Which model scores highest on VQAv2 (val)?
Gemma 3 12B is currently ranked first with 71.6%.
What does VQAv2 (val) measure?
VQAv2 is a balanced Visual Question Answering dataset containing open-ended questions about images that require understanding of vision, language, and commonsense knowledge to answer. VQAv2 addresses bias issues from the original VQA dataset by collecting complementary images such that every question is associated with similar images that result in different answers, forcing models to actually understand visual content rather than relying on language priors.
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.