Skip to main content

Typical LLM Apps with UI

Here are some web based applications built using streamlit.

About how to enable streamlit on KubeAGI, refer to Run LLM application using Streamlit

Simple chat application - English teacher​

  • application definition: refer to config/samples/app_llmchain_englishteacher.yaml
  • application code: examples/app-ui-using-streamlit/chat_app_english_teacher.py
  • application name: base-chat-english-teacher
  • application namespace: kubeagi-system

See details about how it works

Chat application using private knowledgebase​

  • application definition: refer to config/samples/app_retrievalqachain_knowledgebase.yaml
  • application code: examples/app-ui-using-streamlit/chat_using_private_knowledgebase.py
  • application name: base-chat-with-knowledgebase
  • application namespace: kubeagi-system

Here is the workflow described using a chart.

图 1

You should update the application definition/code/name/namespace as your requirement, and then just copy the python script to the streamlit application folder and play with it.

Test Application using curl tool​

KubeAGI APIs can be accessed using <kubeagi-host>/kubeag-apis, and chat mode application can be invoked at /chat context path.

So you can use curl tool to chat with application by using similar command below:

curl -k -XPOST https://portal.172.40.20.125.nip.io/kubeagi-apis/chat --data '{"query":"本制度的适用范围包括哪些人员？","response_mode":"blocking","conversion_id":"","app_name":"kaoqin-bge-qwen-model-app", "app_namespace":"your-namespace"}'

The chat API support blocking and streaming mode, you can use the mode you prefer.