Allow how many number of sent messages attached per request
Is your feature request related to a problem? Please describe.
Sometime, when I ask about something related to each other but don't need LLM to read all the history, I don't want to create a new chat to reduce the input token. Also I don't want to copy/paste a system prompt for every chat.
Describe the solution you'd like
I think we should have a function to allow user to choose how many message they want to send to OpenAI. It should be processed on frontend before sending to server. Let's say I use 2, it will select the last 2 messages then send to API
Originally created by @devanenWl on GitHub (Feb 8, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/9630
# Feature Request
Allow how many number of sent messages attached per request
**Is your feature request related to a problem? Please describe.**
Sometime, when I ask about something related to each other but don't need LLM to read all the history, I don't want to create a new chat to reduce the input token. Also I don't want to copy/paste a system prompt for every chat.
**Describe the solution you'd like**
I think we should have a function to allow user to choose how many message they want to send to OpenAI. It should be processed on frontend before sending to server. Let's say I use 2, it will select the last 2 messages then send to API
<!-- gh-comment-id:2644419332 -->
@atgehrhardt commented on GitHub (Feb 8, 2025):
Looks like we have a function on the community site that does this: https://openwebui.com/f/keke/context_manager/
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @devanenWl on GitHub (Feb 8, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/9630
Feature Request
Allow how many number of sent messages attached per request
Is your feature request related to a problem? Please describe.
Sometime, when I ask about something related to each other but don't need LLM to read all the history, I don't want to create a new chat to reduce the input token. Also I don't want to copy/paste a system prompt for every chat.
Describe the solution you'd like
I think we should have a function to allow user to choose how many message they want to send to OpenAI. It should be processed on frontend before sending to server. Let's say I use 2, it will select the last 2 messages then send to API
@atgehrhardt commented on GitHub (Feb 8, 2025):
Looks like we have a function on the community site that does this: https://openwebui.com/f/keke/context_manager/