Welcome to Demeterics Documentation
Demeterics is an OpenAI-compatible LLM reverse proxy with automatic usage tracking, credit billing, and BigQuery analytics. Track every interaction across Groq, OpenAI, Anthropic, and Gemini—all through a single API.
What is Demeterics?
- Unified API: One API key for Groq, OpenAI, Anthropic, and Gemini
- Automatic Tracking: Every request logged to BigQuery for analytics
- Credit Billing: Pay-per-token via Stripe or bring your own vendor keys
- OpenAI SDK Compatible: Works with existing OpenAI SDKs—just change the base URL
- AI Chat Widget: Embeddable chat widget for customer-facing sites
- Export & Analytics: Download interaction data or query directly in BigQuery
Quick Start
- Get Started in 2 Minutes - Make your first Groq API call
- API Reference - Full endpoint documentation
- Authentication - API keys and security
Use Cases
1. Track All Your LLM Usage
Route all LLM API calls through Demeterics to automatically log:
- Request/response data
- Token usage and costs
- Model performance metrics
- User behavior patterns
2. Single API for Multiple Providers
Use one Demeterics API key to access:
- Groq: Fast, cost-efficient inference
- OpenAI: GPT-4, GPT-4o, embeddings, Whisper
- Anthropic: Claude Sonnet, Opus
- Gemini: Google's latest models
3. Bring Your Own Key (BYOK)
- Store your vendor API keys in Demeterics
- Get full tracking without credit charges
- Maintain compliance while using your own billing
4. Embed AI Chat on Your Website
- AI Chat Widget - Drop-in chat widget
- Domain-restricted for security
- Stripe credit billing or BYOK
- Full conversation tracking
Core Features
OpenAI-Compatible Reverse Proxy
# Just change the base URL - everything else stays the same
curl -X POST https://api.demeterics.com/groq/v1/chat/completions \
-H "Authorization: Bearer dmt_your_api_key" \
-H "Content-Type: application/json" \
-d '{"model": "llama-3.3-70b-versatile", "messages": [{"role": "user", "content": "Hello!"}]}'
BigQuery Analytics
All interactions automatically stored in BigQuery with:
- Full request/response payloads
- Token counts and cost data
- Custom metadata (app, flow, variant, version)
- Timestamp and user tracking
Credit Billing
- Pre-purchase credits via Stripe
- Pay-per-token based on vendor pricing
- Real-time credit balance tracking
- Low balance email alerts
Prompt Engineering Tools
- Triple-slash comments (
///) - Add documentation without token costs - Metadata headers - Track A/B tests, versions, and flows
- Custom analytics - Query by app, flow, variant, environment
Documentation Index
Getting Started
- Quick Start - Your first API call in 2 minutes
- Authentication - API keys and security basics
API & Integrations
- API Reference - Complete endpoint documentation
- SDKs - Python, Node.js, Go, and more
- AI Chat Widget - Embeddable chat for websites
Best Practices
- Prompt Engineering - Metadata, comments, and A/B testing
Supported Providers
| Provider | Base URL | Models |
|---|---|---|
| Groq | https://api.demeterics.com/groq/v1 |
Llama 3.3, Mixtral, Gemma |
| OpenAI | https://api.demeterics.com/openai/v1 |
GPT-4o, GPT-4, embeddings, Whisper |
| Anthropic | https://api.demeterics.com/anthropic/v1 |
Claude Sonnet 4.5, Opus, Haiku |
| Gemini | https://api.demeterics.com/gemini/v1 |
Gemini 2.0 Flash, Pro |
Need Help?
- Support: support@demeterics.com
- Documentation Issues: Report at GitHub Issues
- Feature Requests: Contact us at support@demeterics.com
Ready to get started? Head to the Quick Start Guide to make your first API call.