Getting Started

This guide explains the core objects and execution flow of the Acklix platform. It is intended for developers integrating Acklix via APIs or interfaces.

Core Objects

  • Model

    A model defines inference behavior and configuration. Models are not executable until deployed.

  • Deployment

    A deployment is a versioned, executable instance of a model. All inference requests are routed to a deployment, not directly to a model.

  • Conversation

    Conversations represent persistent interaction state. All messages and executions occur within a conversation context.

  • Inference

    Inference executes input against an active deployment and returns output, tool calls, and metadata.

  • Tool

    Tools are callable functions exposed to a model during inference. Tools may require secrets, services, or dependencies.

  • Workspace

    A workspace isolates models, deployments, tools, conversations, secrets, services, and usage under a single API key.

Execution Flow

  1. Create a model using the Models API.
  2. Deploy the model to create an executable deployment.
  3. Create a conversation under the model.
  4. Send inference requests referencing the deployment and conversation.
  5. Receive model output, tool calls, and metadata.

Authentication

All Acklix APIs are authenticated using an API key. Access tokens are not supported.

Authorization: Api-Key YOUR_API_KEY