@Ueiyester The model sucks baseline which is what i was checking. An open model without a sys prompt shouldn’t be giving garbage ‘policy refusals and canned responses’ on chit chat. The guardrails are heavily baked into the RL / RLHF process. Try the same thing with DeepSeek 4 and GLM 5.2
cited on: glm-5-2
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.