Claude Fable 5.2 tops the Remote Labor Index at 41.3%, nearly double the next model

Claude Fable 5.2 is #1 on the Remote Labor Index, fully automating 41.3% of its projects. That is nearly double the next model, Claude Opus 5.5, at 21.25%. ArtificialWatch read the scores on Scale AI's public leaderboard at 14:50 UTC on November 10.

THE LEADERBOARD

• Claude Fable 5.2: 41.3%
• Claude Opus 5.5: 21.25%
• GPT-6 Astra: 20.83%
• Claude Fable 5.1: 17.92%
• Gemini 4 Argon: 17.08%
• Claude Fable 5: 15.83%

Fable 5.2 more than doubles Fable 5.1, the model it replaces.

WHAT THE TEST MEASURES

The Center for AI Safety and Scale AI built the index from real remote-work projects, including game development, product design, architecture, data analysis and video animation. Some projects cost over $10,000 and took human professionals more than 100 hours. Together they add up to over 6,000 hours of work valued at over $140,000, with every cost and completion time coming from the professionals who did the work. The score is the share of projects an AI agent completes to the point of full automation.

WHAT THE BENCHMARK'S MAKERS SAY

The Center for AI Safety posted on X that when the index launched a year ago, the best AI could fully automate 2.5% of the projects. It says the automation rate is doubling about every three months. The index's own site puts the pace at doubling every 3.3 months, measured over the last three months.

WHAT WE DO NOT KNOW

How Fable 5.2 scores in its fast mode or at lower effort settings, and how Mythos 5.2 or Gemini 4 Carbon would score. Neither appears on the leaderboard.