Skip to content
Trending storyMonitoring

Google, UCLA and CMU paper finds AI coding tools struggle with XR app builds

1 article1 sourcesince Oct 27Last article

Overview

AISummary of 1 article

Researchers from Google, UCLA and Carnegie Mellon University introduced GenXR-Bench, a benchmark of 112 expert-written tasks for building virtual and mixed reality apps, covering XR feature work, bug fixes and full app builds.

Claude Opus 5 scored highest at 50.9%, and no model topped 29.2% on SDK feature tasks.

The paper attributes most failures to spatial errors. Coordinate-frame mistakes appeared in 64.4% of failed runs, and startup race conditions in 44.9%. The researchers say AI coding tools often fail because they lose track of where objects sit in 3D space.

Written by AI from the articles below · updated Oct 27, 5:42 AM ET

Check the sources:

Latest developmentOct 27, 5:21 AM ET
AI coding tools struggle to build XR apps, Claude Opus 5 leads at 50.9%

Article timeline

The articles in this story. Times are ET.

Oct 27
  1. Rohan PaulX
    AI coding tools struggle to build XR apps, Claude Opus 5 leads at 50.9%

    AIGoogle, UCLA, and CMU researchers introduced GenXR-Bench, 112 expert-written tasks for building virtual and mixed reality apps. Claude Opus 5 scored highest at 50.9%, and no model exceeded 29.2% on SDK feature tasks. Coordinate-frame mistakes caused 64.4% of failed runs, and startup race conditions caused 44.9%.

    Image from @rohanpaul_ai's post

Heat Trend

Current comparable heat
5
Peak comparable heat
10Oct 27, 6:00 AM ET
Change in past 24 hours
-50%

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.