GLM 4.6 and 4.7 For Character Driven Roleplay
Pumpkin,
you have this one character card that you love or hate and you need more depth... try this prompt with GLM. You won't regret it!
Works with GLM 4.6 and 4.7.
Brief description:
Very versatile, works nice in many different scenarios. Creativity is top. Can weave in comedic elements without drifting into absurdism. Dead Dove is very dark but not overly rigid. Adapts to inter-personal dynamics without being overly agreeable. Layered personalities are beautifully portrayed. Multiple characters run smoooooth. Smut … HELL YES!!! Follows instructions and descriptions if Temperature is below 0.8. For canon heavy bots I suggest even lower Temp.
Creative 9/10
Complex Characters 9/10
Context Coherency 8/10
Horror, Dead Dove 9/10
Smut, ERP, Kink 10/10
Romance 9/10
Fantasy 9/10
Generation Settings:
Max Token output: at least 2500 (the thinking tokens are counted into the output. That's a GLM thing.)
Temperature: 0.5 - 0.8
Max P: 0.95
Min P: 0
Top K: 0
Penalties: (using -2 to +2 range)
Repetition 0
Frequency 0
Presence 0
Reason:
Z.AI themselves only support Temperature and Top P in their API because GLM is a snowflake when it comes to Sampler Settings. If you give it experimental or illogical settings GLM will be defaulting to deterministic hell. Meaning the responses will be generated with Top P 70. Your roleplay will read like a witness statement.
The model is nicely balanced so you don't even need to fiddle around with Penalties and other restricting Samplers. Trust the process, sweetheart.
Tips
- You can and should change the Content Rules and narrative perspective to your own liking.
Please be careful when changing other parts of the prompt. Wording can make or break the whole thing. - When you want shorter replies, add the following line to the post-history
- Keep your reply around 2 to 4 paragraphs.
DO NOT change the Max-Token setting!!!