Loading page…
Pitch
Description
Comments
Project
llmterm
Your friendly local LLM terminal companion.
5likes2comments
Project details
About this project
LLMTerm enhances your terminal experience by keeping track of your shell commands and providing intelligent suggestions. Setup is easy with automatic model downloading and your data never leaves your computer.
LLMTerm is your ultimate terminal buddy designed to enhance your command-line experience with the power of machine learning. With llmterm, this tool maintains a comprehensive history of your shell usage and offers smart suggestions to streamline your workflow.
Key Features
- Intelligent Suggestions: LLMTerm remembers previous commands, helping you navigate your terminal more efficiently by providing context-aware suggestions.
- CUDA Support: Out of the box, LLMTerm supports CUDA for enhanced performance. You can also customize inference methods by modifying the
Cargo.tomlfile and rebuilding the application.
Building the Project
To get started with LLMTerm, simply run the following command to build the project:
cargo build --release
Exiting the Terminal Buddy
To exit LLMTerm, you can use Ctrl-C, or simply type exit, or quit.
Embrace the future of terminal usage with LLMTerm and enjoy a seamless command-line experience!
Comments
2The underlying library I'm using for inference, kalosm, does support using remote models via API. I've added it to the todo list, and will look into it. Thanks for the suggestion and kind words!
Cool project, should there be a way to use external LLMs (and not just rely on self hosted models?). Although that would probably imply a lot of prompt polishing (making sure not to leak secrets etc).