Documentation

Technical Specifications

Complete integration guide for the Nexus Gateway Protocol. High-performance AI inference with semantic caching.

01.Implementation

Authentication

Generate a unique API Key from the Nexus Dashboard to authenticate your requests.

Configuration

Select your preferred model engine (GPT-4, Claude, Llama 3, or Gemini).

Integration

Authorize requests using the Bearer token protocol in your HTTP headers.

02.Provider Key Pass-Through

Nexus Gateway supports a zero-knowledge key architecture. To utilize your own provider billing and bypass Nexus credit limits, include these headers.

# Set Headers for Sovereign Routing

Authorization:

Bearer nk-your-api-key

# Optional: BYOK Headers

x-nexus-openai-key: sk-your-key

x-nexus-groq-key: gsk_your-key

03.API Reference

POST/api/chat/stream

Primary streaming endpoint for real-time inference. Requests are automatically routed based on the model parameter.

{
                        "model": "llama3-70b-versatile",
                        "message": "Technical query here..."
                      }

# Request Header

Authorization: Bearer nk-your-api-key

GET/api/stats

Returns real-time telemetry including cache efficiency and cost mitigation data.

04.Supported Model Engines

GPT-4o

OpenAI

Llama 3

Groq

Gemini 1.5

Google

Claude 3.5

Anthropic