WorldBench

What is WorldBench?

WorldBench evaluates multimodal large language models on visually diverse reasoning questions, with accuracy as the metric on a curated dataset.

Released

2026-06-04

Evaluates

General AI, Multimodal, Reasoning

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.06538

WorldBench paper

• Paper https://arxiv.org/abs/2606.06538
• Website https://worldbench-vl.github.io/

Which sources cite WorldBench?

• WorldBench

Related General AI, Multimodal, Reasoning benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A-Bench
• A-OKVQA
• A3S-Bench
• AARRI-Bench
• ActBench