Release Notes - Spring Boot Shell LLM Chat v1.0.0
This is the initial release of the LLM Chatbot Spring Boot, bringing the power of Large Language Models directly to your command line!
Key Features:
- Interactive LLM Chat: Engage in conversational dialogues with a configured Large Language Model via CLI
- Simple Command-Line Interface: Easy-to-use commands for initiating and continuing chat sessions.
- Configurable LLM Integration: Supports integration with various LLM providers (implementation details will be in the documentation).
- Spring Boot Shell Powered: Built with Spring Boot Shell, providing a robust and extensible command-line application framework.
Improvements in this Release:
- Initial implementation of the core LLM CLI Spring Boot Chat functionality.
Bug Fixes in this Release:
- No known bugs at the time of this initial release.
Known Issues:
- Limited error handling in certain scenarios.
- Support LLM Model that is compatible with llamacpp
Getting Started:
Using Maven
- Clone this repository
- Download LLM GGUF model from huggingface or other source
- Use mvn package to build the module into jar file
mvn clean package- The following command should be used to run the Java application
java -Dllamacpp.model=<path to gguf model> -jar <path to jar that is build from maven>Using Docker
- Clone this repository
- Download LLM GGUF model from huggingface or other source, and place the downloaded model in the "model" folder.
- Run docker build
docker build -t chat-cli .- Run the docker image from previous step:
docker run -v <path to gguf file>:/app/llm-model.gguf -it chat-cli