llm cli

LLM CLI

A CLI tool to interact with LLMs. Provide your API keys in this format: <completer>_API_KEY>

The supported completers are:

• ANTHROPIC
• MISTRAL
• OPENAI
• PERPLEXITY

Installation

Currently, you can build from source and use the generated binary. Clone the repository and run the following command:

make

This will create a binary named llm in the bin directory.

Features ?

1. Context persistence

All interactions are stored locally on your filesystem. This setup allows for multiple prompts to be sent to LLMs, with context being maintained across different conversations.

2. Shared conversational history

Your conversation history can be shared among all supported LLMs. This allows you to leverage the strengths of each model, combining their capabilities to create a powerful conversational experience.

3. Terminal assistance

Prefixing any command with llm provides a clear and concise explanation of that command, along with examples demonstrating its usage.

4. Multiple sessions

You can have multiple sessions, each with their own context and history. This allows you to switch between different conversations and maintain a consistent conversational experience.

Coming soon ?

• Default models for each variant
• Carefully crafted system prompts for terminal assistance
• Multiple sessions
• UI using tview

Usage

1. CLIs

To use the CLI, you need to provide your API keys as environment variables. You can set these variables in your shell configuration file (e.g., .zshrc) or directly in the terminal.

Once you've set the environment variables, you can use the following command (from repo root) to prompt the LLM:

bin/llm <llm-name> [flags] <prompt>

To chat, simply use the chat subcommand:

bin/llm <llm-name> chat

And of course, add /path/to/repo/bin to your $PATH to use the CLI from anywhere.

2. Command assist

Create an environment variable DEFAULT_COMPLETER with the name of the completer you want to use. They are:

• claude
• mistral
• gpt

Then, you can prefix any command you want to understand with llm to understand how it works with examples.