INSTALL.md

Installation Guide for llm-chutes

Prerequisites

1. Make sure you have Python 3.9+ installed
2. Install the LLM CLI tool if you haven't already:

   pip install llm

Installation

Option 1: Install from source (recommended for development)

1. Clone or download this repository
2. Navigate to the project directory
3. Install in development mode:

   pip install -e .

Option 2: Install using pip (when published)

pip install llm-chutes

Configuration

1. Get your API key from Chutes AI

2. Set it as an environment variable:

   export CHUTES_API_KEY="your-api-key-here"

   Or add it to your shell profile (.bashrc, .zshrc, etc.):

   echo 'export CHUTES_API_KEY="your-api-key-here"' >> ~/.zshrc
   source ~/.zshrc

   Alternatively, you can use the LLM keys system:

   llm keys set chutes

Verification

1. Test the installation:

   python3 test_basic.py

2. List available models:

   llm chutes models

3. Try a simple prompt:

   llm -m chutes/deepseek-ai/DeepSeek-R1 "Hello, how are you?"

Troubleshooting

"No models found" error

• Make sure your CHUTES_API_KEY is set correctly
• Try refreshing models: llm chutes refresh
• Check your internet connection

Plugin not recognized

• Make sure you installed in the same Python environment as LLM
• Try reinstalling: pip uninstall llm-chutes && pip install -e .

API errors

• Verify your API key is valid
• Check if the Chutes AI service is accessible
• Try the test script: python3 test_basic.py

Usage Examples

# List all models
llm chutes models

# List models in JSON format
llm chutes models --json

# Refresh model cache
llm chutes refresh

# Use a model with caching
llm -m chutes/deepseek-ai/DeepSeek-R1 -o cache 1 "Explain quantum computing"

# Set up an alias for easier use
llm aliases set deepseek chutes/deepseek-ai/DeepSeek-R1
llm -m deepseek "What is machine learning?"