API Integration

This guide explains how to integrate external applications with Acklix using the REST API.

Overview

Acklix exposes a REST API for managing models, deployments, conversations, and running inference.

All execution requests are routed through deployments. Models cannot be used directly for inference.

Authentication

All API requests require an API key. Access tokens and OAuth-based authentication are not supported.

Authorization: Api-Key YOUR_API_KEY

API keys are scoped to a workspace and must be kept server-side.

Base URL

https://platform.acklix.com

Running Inference

To execute a model, send a request to the inference endpoint using a deployment ID.

POST /api/v1/inference

Required Fields

  • deployment_id — Active deployment identifier
  • input — Input text or payload

Example (cURL)

curl -X POST "https://platform.acklix.com/api/v1/inference" \
  -H "Authorization: Api-Key YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "deployment_id": "dep_12345",
    "input": "How can I reset my password?"
  }'

Identifiers

  • Model ID — Identifies a model configuration
  • Deployment ID — Identifies an executable model instance
  • Conversation ID — Identifies a persistent interaction

Only deployment IDs are accepted by the inference API.

Error Handling

Errors are returned using standard HTTP status codes with a structured error payload.

{
  "error": "deployment_not_found"
}

Common Errors

  • 401 — Invalid API key
  • 404 — Deployment not found
  • 409 — Deployment not active
  • 429 — Rate limit exceeded