Specialist test. Evaluate this separately from the eight core tests. Explore the active suite →
Focused model compare
Yingzao Fashi Assembly
Claude Opus 5.5 is the starting run, compared with GPT-6 Astra, GPT-6 Sol, DeepSeek V4.1 Flash.
Starting model
Claude Opus 5.5
OpenAI
GPT-6 Astra
OpenAI
GPT-6 Sol
DeepSeek