# The Ethical AI Company — System & API Specification

100% Private, Carbon-Neutral AI Gateway & Infrastructure.
100% Private & Operated • Zero Data Retention • 100% Renewable Geothermal & Hydroelectric Compute.

## API Endpoint
- Base URL: `https://ethicalaico.com/v1`
- Format: OpenAI-compatible (`/v1/chat/completions`, `/v1/models`)
- Authorization: Header `Authorization: Bearer YOUR_API_KEY`

## Model Lineup & Auto-Router

### 1. Auto-Router (`ethical-ai-auto`) — Recommended Default
Dynamically classifies prompt intent and dispatches to the optimal backend engine:
- Code & Mathematics -> `ethical-ai-pro` (DeepSeek V4 Pro)
- Creative & Short Chat -> `ethical-ai-ultra` (Gemini 3.6 Flash)
- Long Context (>= 800 tokens) -> `ethical-ai-flash` (DeepSeek V4 Flash)
- Image Requests -> 3-Tier Nano Banana Generative Engine

### 2. Discrete Model IDs
- `ethical-ai-flash`: DeepSeek V4 Flash (~1.2s latency, $0.22/M in, $0.66/M out off-peak)
- `ethical-ai-pro`: DeepSeek V4 Pro (~3.5s latency, $0.66/M in, $1.98/M out off-peak)
- `ethical-ai-ultra`: Gemini 3.6 Flash (~0.35s latency, $1.50/M in, $7.50/M out)

## Audit & Cost Telemetry Headers
Every response includes custom headers:
- `X-EthicalAI-Route-Reason`: Explanation of model selection decision.
- `X-EthicalAI-Cost-USD`: Real-time blended upstream cost in USD.
- `X-EthicalAI-Power-Source`: `100% Renewable Geothermal & Hydroelectric`
- `X-EthicalAI-Data-Retention`: `Zero Logs / Ephemeral Memory`

## Quick Start Example (Python)
```python
import openai

client = openai.OpenAI(
    base_url="https://ethicalaico.com/v1",
    api_key="eai_live_YOUR_KEY"
)

response = client.chat.completions.create(
    model="ethical-ai-auto",
    messages=[{"role": "user", "content": "Write a python function to parse JSON"}]
)

print(response.choices[0].message.content)
```

## Contact & Enterprise
- Founder: The Ethical AI Company
- Support: support@ethicalaico.com
- Web App: https://ethicalaico.com/app