Selecting a character on SpicyChat opens the most recent conversation with that character. Conversations are automatically saved, so you can return to them later from the chat menu instead of starting over each time.
This makes saved chats part of the normal conversation flow: you can reopen an existing conversation, start a separate one when you want a fresh thread, and use the available chat controls while the earlier conversation remains saved.
Start a new chat or return to a saved conversation
If you select a character you have already chatted with, SpicyChat opens the most recent conversation with that character. To begin a separate conversation instead, you can start a new chat from the conversation menu. The conversation you were already using remains saved.
The Chats menu lists your conversations and lets you continue the most recent chat. Because conversations are saved automatically, you do not need to manually save a conversation before leaving it.
Use roleplay actions and understand reply length
SpicyChat supports a simple roleplay action format inside messages. You can place an action between asterisks, and the character can respond to that action as part of the conversation.
Character responses are also measured in tokens. Tokens are units of text processed by the AI and can include whole words, parts of words, symbols, and emoji.
Character replies are described as roughly 120-180 tokens. The maximum reply-token limit depends on the SpicyChat plan:
| Plan | Maximum reply tokens |
|---|---|
| Free | 180 |
| Get A Taste | 180 |
| True Supporter | 300 |
| I’m All In | 300 |
The 120-180-token figure describes character replies generally, while the plan values above are maximum reply-token ceilings. They are related measures, but they are not the same thing.
Report an inappropriate response
If a character produces an inappropriate response, you can report that response. The reported response is flagged for moderation and removed from that conversation.
If you need additional help beyond the in-chat report action, SpicyChat also provides Discord for real-time assistance and a support-ticket route. You can use the available SpicyChat support options for account or service issues that need further assistance.
Can character creators read your chat logs?
No. The SpicyChat chat guide states that character creators cannot view users' chat logs. This rule specifically describes visibility for the person who created the character and should not be expanded into a broader claim about every platform or moderation function.
Understand how Lorebook context can affect a chat
Lorebooks can influence an ongoing conversation when their keywords appear in recent messages. If a Lorebook keyword appears in the last four messages from either the user or the character, the corresponding entry content can be added to the chat context for the next response.
Once an entry is activated, it stays active for about two turns. Repeating the relevant keyword can extend that active period.
This is a conversation-level effect of a configured Lorebook rather than a permanent unlimited-memory function. To understand entries, keywords, attachment limits, and the supported character scope, you can learn about Lorebooks.
Questions about SpicyChat chats
Are SpicyChat chats saved automatically?
Yes. Conversations are automatically saved and can be viewed from the chat menu.
How do I start a new chat?
You can start a new chat from the conversation menu. The conversation you are currently using remains saved when you start the new one.
How long can a character reply be?
Character replies are described as roughly 120-180 tokens. The maximum is 180 tokens on Free and Get A Taste, and 300 tokens on True Supporter and I’m All In.
How do I report an inappropriate response?
You can report the response in the chat. It is then flagged for moderation and removed from that conversation.
Can a character creator read my chat logs?
No. SpicyChat's chat guide states that character creators cannot view users' chat logs.
What triggers Lorebook context in a chat?
A Lorebook entry can be triggered when one of its keywords appears in the last four messages from the user or character. Its content can then enter the context for the next response. An activated entry stays active for about two turns, and repeated keyword use can extend that period.
This page explains published service information and is not official account support. Features and policies can change.