Vision & Language Benchmark

WorldVQA (with tools)

WorldVQA visual question answering evaluated with tools, as printed on the Muse Spark Open model card (xhigh reasoning effort). Distinct from RealWorldQA.

Category

Vision & Language

Score type

percent

Scale

100%

higher is better

Models tested

1

What is WorldVQA (with tools)?

WorldVQA (with tools) is a benchmark for understanding images combined with text. The official description says: "WorldVQA visual question answering evaluated with tools, as printed on the Muse Spark Open model card (xhigh reasoning effort)". Below we describe in simple words what benchmarks of this type generally examine.

What does it test?

WorldVQA (with tools) measures understanding images combined with text. In detail, it checks:

  • answering questions about images (figures, photos, charts, memes)
  • reading text inside images and connecting it to the question
  • visual reasoning such as counting, locating and comparing objects
  • combining what is seen with general knowledge

How to read the score

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: WorldVQA visual question answering evaluated with tools, as printed on the Muse Spark Open model card (xhigh reasoning effort). Distinct from RealWorldQA.

Frequently asked questions

What is the WorldVQA (with tools) benchmark?

WorldVQA (with tools) is a benchmark for understanding images combined with text. The official description says: 'WorldVQA visual question answering evaluated with tools, as printed on the Muse Spark Open model card (xhigh reasoning effort)'. Below we describe in simple words what benchmarks of this type generally examine.

What does WorldVQA (with tools) test?

WorldVQA (with tools) mainly measures understanding images combined with text. Concretely, it checks: answering questions about images (figures, photos, charts, memes); reading text inside images and connecting it to the question; visual reasoning such as counting, locating and comparing objects.

How is the WorldVQA (with tools) score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get WorldVQA (with tools) data and scores?

The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/worldvqa-with-tools/scores) or pull it directly from the Open LLM Radar API with any HTTP client.

Model scores 1 model tested

Reset
# Model Score Date Source Verified
1 Muse-Spark-Open-816B-A42B
Meta Superintelligence Lab
49.4%
15.10.2026 Report Verified