Skip to content

Artificial Analysis Reports Four Models Added Sagrada Família’s Magic Square to AA-3D Scenes

1 reports 1 reporting sources 1 hours agoNew

Understand the event

AI overview

Artificial Analysis reported that four models unexpectedly added the Sagrada Família’s magic square to their AA-3D scenes while offline and without being prompted to include it. Claude Fable 5.1 and Sonnet 5.5 displayed the square visibly, while Grok 4.7 and Step 5 Preview placed it behind a pillar.

Generated by AI from reports · Updated 46 minutes ago

Models in this story

  • Claude Fable 5.1#4 overall$10 input / $50 output per million tokens
  • Grok 4.7#12 overall$2 input / $6 output per million tokens
  • Step 5 Preview#18 overall$1 input / $2.7 output per million tokens

Report timeline

Follow the reports to explore different sides of the event.

10/22
  1. Four Models Unexpectedly Added the Sagrada Família’s Magic Square to AA-3D Scenes

    Four models unexpectedly added the Sagrada Família’s magic square to their AA-3D scenes, offline and unprompted: Claude Fable 5.1 and Sonnet 5.5 featured it visibly, whereas Grok 4.7 and Step 5 Preview hid it behind a pillar The sculptor Josep Maria Subirachs carved the square https://t.co/HI1DZHXbwE

Tomorrow at 08:00, a briefing like this arrives in your inbox.

The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue

Event heat trend

There is not enough continuous observation data to plot a trend yet.

Keep reading

Latest daily · Issue 28Mistral AI releases Voxtral 3 speech models and a voice agent SDK

Trending now

  1. 1Mistral and NVIDIA Announce Industrial Engineering AI Collaboration and Open Benchmarks
  2. 2Mistral Releases Voxtral 3 Speech Models, Voice Agents API and LiveKit-Based SDK
  3. 3PHP releases security updates after OpenAI says its models helped researchers find critical flaws
  4. 4DeepSeek Open-Sources DeepInfer and API Components, With Model and Hardware Limits