Developer Reference

Roboflow provides a comprehensive set of developer tools for building, deploying, and managing computer vision models. This reference section covers all the SDKs, CLIs, APIs, and infrastructure components you need.

Tools Overview

Roboflow's developer ecosystem includes several components that work together:

  • Python SDK -- The Roboflow Python SDK for managing workspaces, projects, datasets, and training jobs.
  • Inference SDK -- Lightweight Python client for communicating with the Inference Server.
  • Inference CLI -- Command-line tool for managing the Inference Server and running common tasks.
  • Roboflow CLI -- Command-line interface for interacting with the Roboflow platform.
  • REST API -- HTTP API for programmatic access to the Roboflow platform.
  • Inference Server -- Core Python package and HTTP server for model loading, inference, and Workflows execution.

Inference Architecture

The diagram below shows how the inference components fit together.

  • inference-sdk -- Lightweight Python client for communicating with the Inference Server.
  • inference-cli -- Command-line tool for managing the Inference Server and running common tasks.
  • Inference Server -- HTTP server (Docker) that wraps the inference package as a REST API.
  • inference -- Core Python package for model loading, inference, and Workflows execution.

Authentication

Before using any of the developer tools, you will need to authenticate. See the Authentication section for details on finding and managing your API keys.

Account Management

  • Workspace -- Create and manage workspaces, team members, and roles.
  • Billing -- Understand credits, plans, and billing.
  • Support -- Get help from the Roboflow support team.

Rapid

  • Roboflow Rapid -- Go from raw data to a ready-to-use object detection API in five minutes.