llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeBenchmarksmultimodalWorldVQA

multimodal benchmark

WorldVQA

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

Published models5
Registry coverage5
MetricScore
EvidenceB

On this page

  • Ranking
  • Distribution
  • Highlights
  • About
  • FAQ

WorldVQA leaderboard

Sorted by the source-provided rank. Higher score is better according to the registry.

5 rows
Columns

Show columns

1ACQwen3.7-PlusAlibaba Cloud / Qwen Team61.1%100.0%5CAug 7, 2026
2BYSeed 2.1 ProByteDance53.0%75.0%5CAug 7, 2026
3MAKimi K3Moonshot AI51.0%50.0%5CAug 7, 2026
4BYSeed 2.1 TurboByteDance48.6%25.0%5CAug 7, 2026
5MAKimi K2.5Moonshot AI46.3%0.0%5CAug 7, 2026

Score distribution

Top published rows on the benchmark's original scale.

WorldVQA

WorldVQA highlights

The top published results on this benchmark's own scale.

Rank #1Qwen3.7-Plus61.1%Rank #2Seed 2.1 Pro53.0%Rank #3Kimi K351.0%Rank #4Seed 2.1 Turbo48.6%

What is WorldVQA?

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.

Family
WorldVQA
Modality
multimodal
Primary category
multimodal
Score direction
higher
LLMBoard eligible
No
Evaluation key
worldvqa|llm-stats-current

Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.

FAQ

Common questions about WorldVQA.

Which model scores highest on WorldVQA?

Qwen3.7-Plus is currently ranked first with 61.1%.

What does WorldVQA measure?

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.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

5 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.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai