Prerequisites
Please answer the following questions for yourself before submitting an issue.
Expected Behavior
Using the multi-user feature, two people who uses the same koboldcpp backend should only have the reply for their character, and not another character's reply.
Current Behavior
A person can have the reply of another person's character instead of their character if they share the same koboldcpp backend with the multi-user feature activated.
Environment and Context
Please provide detailed information about your computer setup. This is important in case the issue is not reproducible except for under certain specific conditions.
- Physical (or virtual) hardware you are using, e.g. for Linux:
$ lscpu
- Operating System, e.g. for Linux:
$ uname -a
- SDK version, e.g. for Linux:
$ python3 --version
$ make --version
$ g++ --version
Failure Information (for bugs)
Nothing in the logs really show that.
Steps to Reproduce
- Having two persons sharing the same koboldcpp backend (same IP) with multi-user feature activated.
- step 2 (I suppose ?) Sending a request while another person is waiting for their reply.
Prerequisites
Please answer the following questions for yourself before submitting an issue.
https://github.com/ggerganov/llama.cpp/discussions), and have a new bug or useful enhancement to share.
Expected Behavior
Using the multi-user feature, two people who uses the same koboldcpp backend should only have the reply for their character, and not another character's reply.
Current Behavior
A person can have the reply of another person's character instead of their character if they share the same koboldcpp backend with the multi-user feature activated.
Environment and Context
Please provide detailed information about your computer setup. This is important in case the issue is not reproducible except for under certain specific conditions.
$ lscpu$ uname -aFailure Information (for bugs)
Nothing in the logs really show that.
Steps to Reproduce