A user on Reddit documented the evolution of Anthropic's Claude models by giving them the same complex prompt over 14 months. The prompt requested the creation of a 3D website resembling a video game with procedural terrain, interactive objects, and continuous forward movement. The user observed differences in how each Claude model, including versions of Sonnet and Opus, interpreted and executed the prompt, noting specific issues like a shader bug in Sonnet 5.5. AI
IMPACT Provides insight into the iterative improvements and specific capabilities of different Claude model versions.
RANK_REASON User-generated analysis and comparison of existing models, not a primary release or research paper.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →