Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.geekflare.com/llms.txt

Use this file to discover all available pages before exploring further.

When you interact with an AI model, it does not possess memory like a human does. To help the AI follow your conversation, Geekflare Chat forwards your recent chat history (context) to the LLM alongside your new prompt. How much the AI remembers depends entirely on two factors: the number of previous messages configured in your settings and the maximum context window supported by the LLM you are using.

​
Default Memory Behavior

To provide a good experience while optimizing your costs, Geekflare Chat sends the previous 10 messages to the AI by default. This gives the model enough background to understand ongoing topics without consuming excessive credits.

​
Customizing Your Chat Memory

You have full control over how much history is sent to the AI. By going to Chat Settings, you can adjust the context limit from 0 to up to 50 previous messages. Chat Memory Settings

​
Credit Usage and the Long Context Notice

Please be aware that increasing your context limit comes with a trade-off. Sending a higher number of previous messages requires the AI to process more data, which will trigger higher credit usage for every new message you send. If your active conversation exceeds 20,000 tokens, Geekflare Chat will display a Long Context Notice banner. This is a friendly alert to help you avoid unexpected credit consumption. You can learn more about how this works and how to manage your costs in our Long Context Notice guide.