CacheGPT CLI

Intelligent LLM caching proxy CLI - Save costs and improve response times by caching similar LLM queries.

Installation

npm install -g cachegpt-cli
# or
yarn global add cachegpt-cli

Features

• 🚀 Intelligent Caching: Automatically cache and reuse similar LLM responses
• 💰 Cost Savings: Reduce API costs by up to 70% through smart caching
• ⚡ Faster Responses: Retrieve cached responses in milliseconds
• 🔌 Multiple Providers: Support for OpenAI, Anthropic, and other LLM providers
• 📊 Usage Analytics: Track requests, cache hits, and cost savings
• 🎯 Semantic Search: Find similar queries using vector embeddings
• 🔒 Secure: API key encryption and secure storage

Quick Start

1. Initialize Configuration

cachegpt init

This will guide you through setting up:

• API endpoint URL
• Authentication credentials
• Cache preferences

2. Test Connection

cachegpt test

3. Start Using

# Interactive chat
cachegpt chat

# View statistics
cachegpt stats

# Clear cache
cachegpt clear --older-than 24h

Commands

init

Initialize CacheGPT configuration with interactive setup.

test

Test API connectivity and cache functionality.

cachegpt test
cachegpt test --verbose

stats

Display cache statistics and usage metrics.

cachegpt stats
cachegpt stats --format json
cachegpt stats --days 30

clear

Clear cache entries.

cachegpt clear
cachegpt clear --older-than 24h
cachegpt clear --all

config

Manage configuration settings.

cachegpt config get api-url
cachegpt config set api-url https://api.cachegpt.io
cachegpt config list

chat

Start an interactive chat session with your LLM.

cachegpt chat
cachegpt chat --model gpt-4

Configuration

Configuration is stored in ~/.cachegpt/config.json:

{
  "apiUrl": "https://api.cachegpt.io",
  "apiKey": "cgpt_...",
  "defaultModel": "gpt-3.5-turbo",
  "cacheEnabled": true,
  "logLevel": "info"
}

Environment Variables

You can also use environment variables:

export CACHEGPT_API_URL=https://api.cachegpt.io
export CACHEGPT_API_KEY=cgpt_your_api_key

API Usage

The CLI can be used programmatically:

const { CacheGPT } = require('cachegpt-cli');

const client = new CacheGPT({
  apiKey: 'cgpt_...',
  apiUrl: 'https://api.cachegpt.io'
});

// Send a query
const response = await client.chat('What is machine learning?');
console.log(response.content);

// Check if response was cached
if (response.cached) {
  console.log(`Cache hit! Saved: $${response.costSaved}`);
}

Examples

Basic Chat

$ cachegpt chat
> What is artificial intelligence?

AI is the simulation of human intelligence by machines...
[Cached: No | Response time: 1.2s | Cost: $0.002]

> What is AI?

AI is the simulation of human intelligence by machines...
[Cached: Yes | Similarity: 95% | Response time: 0.05s | Saved: $0.002]

View Statistics

$ cachegpt stats

┌─────────────────┬──────────┐
│ Metric          │ Value    │
├─────────────────┼──────────┤
│ Total Requests  │ 1,234    │
│ Cache Hits      │ 892      │
│ Hit Rate        │ 72.3%    │
│ Total Saved     │ $45.67   │
│ Avg Response    │ 145ms    │
└─────────────────┴──────────┘

Clear Old Cache

$ cachegpt clear --older-than 7d

Cleared 156 cache entries older than 7 days.

Troubleshooting

Connection Issues

If you're having connection issues:

1. Check your API key: cachegpt config get api-key
2. Verify the API URL: cachegpt config get api-url
3. Test connectivity: cachegpt test --verbose

Cache Not Working

1. Ensure cache is enabled: cachegpt config get cache-enabled
2. Check cache statistics: cachegpt stats
3. Clear and rebuild cache: cachegpt clear --all

Support

• Documentation: https://cachegpt.io/docs
• Issues: https://github.com/cachegpt/cachegpt-cli/issues
• Email: support@cachegpt.io

License

MIT © CacheGPT Team