[Feature] support kimi ai
Problem Description cant use kimi ai
Proposed Solution support kimi ai module
Additional Context https://platform.moonshot.cn/docs/intro
It is recommended to refer to the implementation method in this project, which involves integrating refresh token and access token with Kimi.
I suggested that you could conisder to support kimi as the kimi AI model is smart that deepseek and chatgpt on ubuntu+tensorflow+nvidia+docker related issues. Please consider to support kimi AI model in the future, thank you so much.
Let me give a example. I tried to use a docker which is provided by https://blog.mypapit.net/2025/06/how-to-get-tensorflow-acceleration-with-nvidia-rtx-50-series-gpu-with-docker-rtx5060ti-16gb-for-ubuntu-and-windows-wsl2.html and I followed it's step to setup a docker to execute tensorflow and support RTX 5060.
I faced a issue and it's "Loaded runtime CuDNN library: 9.5.1 but source was compiled with: 9.7.1. CuDNN library needs to have matching major version and equal or higher minor version.".
I checked the answer from ChatGPT, DeepSeek and Kimi. Both ChatGPT and Deepseek asked me to reinstall cudnn libraries and Kimi asked me to check what versions of cudnn is insatlled in my os and it's command is "ls -l /usr/lib/x86_64-linux-gnu/ | grep libcudnn".
Kimi is the only AI model that ask me to check this command and it's the only right answer for my problem. I followed the Kimi's suggestion and they are: ls -l /usr/lib/x86_64-linux-gnu/ | grep libcudnn echo $LD_LIBRARY_PATH export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:$LD_LIBRARY_PATH
And above three steps are the only right solution to fix problems. Other AI models don't mention this.
Therefore, I suggested that you could kindly consider to support Kimi. For users work with operating system(Linux, Python, Driver, Tensorflow) and need to handle issues of OS, only Kimi gave right solutin.
hi all,
you can add by yourself. the api host: https://api.moonshot.ai/v1 the api path: keep it empty
then, it works.