Comment by zb3

We can know if this is Anthropic/OpenAI by testing the "guardrails" - absurd guardrails = it's them, reasonable/no guardrails = Chinese models..

(as a bonus - thinking forever = GLM)