# Komilion - AI Model Router & Gateway
# https://www.komilion.com
# Contact: support@komilion.com

## What is Komilion?

Komilion (Chameleon + Million) is an intelligent LLM API gateway that provides:
- One API key for 400+ AI models (OpenAI, Anthropic, Google, Meta, Mistral, etc.)
- Automatic model selection based on task type ("Neo Mode")
- OpenAI SDK compatibility - just change the base URL
- Real-time model benchmarks from LLM Arena and Artificial Analysis
- Cost optimization with intelligent routing
- Pay-per-use pricing with no subscriptions

## Key Features

### Neo Mode (Automatic Model Selection)
Neo Mode analyzes your prompt and automatically selects the best model based on:
- Task type (coding, reasoning, creative writing, etc.)
- Real-time benchmark scores (ELO ratings, speed, reliability)
- Cost policy (frugal, balanced, or premium)

### Supported Providers
- OpenAI (GPT-4, GPT-3.5)
- Anthropic (Claude 3, Claude 2)
- Google (Gemini Pro, Gemini Flash)
- Meta (Llama 3)
- Mistral (Mistral Large, Mixtral)
- And 400+ more models

### Doc Whisperer
Extract structured data from documents (PDFs, images) using AI agents.

## API Quick Start

Base URL: https://api.komilion.com/v1
Authentication: Bearer token (API key from dashboard)

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.komilion.com/v1",
    api_key="your-komilion-api-key"
)

# Use Neo Mode for automatic model selection
response = client.chat.completions.create(
    model="neo-mode/balanced",  # or frugal, premium
    messages=[{"role": "user", "content": "Hello!"}]
)

# Or specify a model directly
response = client.chat.completions.create(
    model="anthropic/claude-3-opus",
    messages=[{"role": "user", "content": "Hello!"}]
)
```

## Pricing

- Pay only for what you use (no subscriptions)
- Prices vary by model (passed through from providers + small margin)
- Neo Mode frugal policy: ~$0.10-0.50 per 1M tokens
- Free trial credits for new users

## Important Pages

- Homepage: https://www.komilion.com
- Documentation: https://www.komilion.com/docs
- Pricing: https://www.komilion.com/pricing
- Model Explorer: https://www.komilion.com/models
- Playground: https://www.komilion.com/playground
- How It Works: https://www.komilion.com/how-it-works

## Rate Limits

- Standard: 120 requests per minute
- After completing Neo School: up to 200 requests per minute
- Rate limit exceeded returns HTTP 429

## Support

- Email: support@komilion.com
- Contact form: https://www.komilion.com/contact

## Legal

- Terms of Service: https://www.komilion.com/terms
- Privacy Policy: https://www.komilion.com/privacy

---
Last updated: 2026-01-01