Skip to main content
BenchmarkComputer Visionv1.0

RealWorldQA

by xAI · open-source · Last verified 2026-03-01

Benchmark testing multimodal models on practical real-world visual understanding tasks. Features questions about real photographs requiring spatial reasoning, object recognition, scene understanding, and practical knowledge that goes beyond simple object detection.

https://huggingface.co/datasets/xai-org/RealWorldQA
C+
C+Average
Adoption: C+Quality: AFreshness: ACitations: C+Engagement: F

Specifications

License
CC-BY-4.0
Pricing
open-source
Capabilities
model-evaluation, real-world-vision-testing, spatial-reasoning-assessment
Integrations
huggingface
Use Cases
visual-understanding-evaluation, practical-vision-testing, real-world-assessment
API Available
No
Evaluated Models
claude-4, gpt-5, gemini-2.5-pro
Metrics
accuracy, spatial-accuracy
Methodology
Multiple-choice questions about real photographs testing spatial reasoning, counting, scene understanding, and practical visual knowledge.
Last Run
2026-02-20
Tags
benchmark, evaluation, multimodal, real-world, visual-understanding
Added
2026-03-17
Completeness
100%

Index Score

53.1
Adoption
58
Quality
82
Freshness
86
Citations
54
Engagement
0

Explore the full AI ecosystem on Agents as a Service