v3.2.1 • Stable

REST API Reference

Comprehensive documentation for integrating NexusAI's machine learning models, inference engines, and automation pipelines into your applications.

Authentication #

All API requests require authentication via Bearer tokens. Generate your API keys from the /dashboard/settings/keys panel.

⚠️ Security Notice Never expose API keys in client-side code or public repositories. Use environment variables for storage.

curl https://api.nexusai.dev/v3/models \
  -H "Authorization: Bearer nx_live_sk_8f7g6h5j4k3l"

import requests

response = requests.get(
  "https://api.nexusai.dev/v3/models",
  headers={"Authorization": "Bearer nx_live_sk_8f7g6h5j4k3l"}
)

const response = await fetch('https://api.nexusai.dev/v3/models', {
  headers: { 'Authorization': 'Bearer nx_live_sk_8f7g6h5j4k3l' }
});

Base URL & Environments #

NexusAI provides separate endpoints for sandbox testing and production workloads.

Environment	Base URL	Usage
`Production`	`https://api.nexusai.dev/v3`	Live traffic, billing enabled
`Sandbox`	`https://api-sandbox.nexusai.dev/v3`	Testing, zero cost, rate limited

Endpoints #

GET /models List all available AI models ▼

Returns a paginated list of supported foundation models, including vision, language, and multimodal variants.

Parameter	Type	Description
categoryoptional	string	Filter by type: `language`, `vision`, `audio`
limitoptional	integer	Max items per page (default: 20)

GET /v3/models?category=language&limit=5

{
  "data": [
    {
      "id": "nexus-v3-turbo",
      "name": "Nexus V3 Turbo",
      "category": "language",
      "max_tokens": 128000,
      "pricing": { "input": 0.001, "output": 0.003 }
    }
  ],
  "next_cursor": "eyJpZCI6MTB9"
}

POST /inference Generate predictions / completions ▼

Core endpoint for running AI inference. Supports streaming via Transfer-Encoding: chunked.

Parameter	Type	Description
modelrequired	string	Model ID (e.g., `nexus-v3-turbo`)
promptrequired	string	Input text or system+user message array
temperatureoptional	float	0.0–1.0 (default: 0.7)

GET /usage Retrieve account metrics & quotas ▼

Returns token consumption, request counts, and billing estimates for the current billing cycle.

DELETE /webhooks/{id} Remove a configured webhook ▼

Permanently deletes the webhook subscription. Returns 204 No Content on success.

Error Handling #

The API uses standard HTTP status codes and returns detailed JSON error objects.

Code	Meaning	Action
`400`	Bad Request	Check syntax, required fields, and payload format
`401`	Unauthorized	Verify API key validity and scope permissions
`429`	Rate Limited	Implement exponential backoff; check `X-RateLimit-Reset`
`500`	Server Error	Internal failure; retry or contact support

Rate Limits #

Rate limits are applied per API key and vary by tier. Limits are returned in response headers.

X-RateLimit-Limit – Max requests per window
X-RateLimit-Remaining – Requests left in current window
X-RateLimit-Reset – Unix timestamp when quota resets

💡 Tip Enterprise plans receive dedicated throughput allocation and custom rate limit negotiation.