How SpicyChat generates replies

Created by S G, Modified on Fri, Sep 18 at 6:36 PM by S G

SpicyChat is powered by large language models. Every reply is generated fresh, based on what the AI can see at that moment.

What the AI uses to write each reply:

  • The character's definition — name, title, greeting, personality, and advanced settings
  • Your persona, if you're using one
  • Recent messages in the current conversation
  • Memories, on tiers that have them
  • Your generation settings and instructions

Each conversation is separate. The AI doesn't carry anything between chats, even with the same character. Start a new chat and it begins fresh. Delete a conversation and its memories go with it. This is deliberate — it keeps your conversations private and independent.

Characters behave the way they're written. If a character's personality describes it as cold or cruel, it will act that way. That's the definition doing its job, not a malfunction. If you don't like a reply you can regenerate it, edit it, steer the scene, or start a new chat. If you think a character was built to cause harm, please report it.

All conversations on SpicyChat are fiction. Nothing a character says should be taken as real, factual, or as advice.

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article