Getting Started with Roboflow

Roboflow provides everything you need to build and deploy computer vision applications: upload data, annotate images, train models, build multi-step Workflows, and deploy anywhere.

What You Can Do

Deployments - Run models in the cloud or on your own hardware.
Workflows - Run multi-stage vision workflows anywhere.
Models - Train state-of-the-art models with a few clicks.
Datasets - Upload, manage, annotate, and analyze images.
Workspaces - Invite, remove, and manage users in your workspace.
Developer Reference - Learn about the Roboflow CLI and Python SDK.

Inference: The Deployment Engine

Inference is Roboflow's open-source computer vision deployment hub. It handles model serving, video stream management, pre/post-processing, and GPU/CPU optimization so you can focus on building your application.

Deploy Anywhere

	Serverless	Dedicated	Self-Hosted
Fine-Tuned & Pre-Trained Models	Yes	Yes	Yes
Workflows	Yes	Yes	Yes
Foundation Models		Yes	Yes
Video Streaming		Yes	Yes
Dynamic Python Blocks		Yes	Yes
Runs Offline			Yes
Billing	Per-Call	Hourly	Free + metered

Serverless - Pay-per-inference, scales to zero. Best for getting started quickly.
Dedicated - Single-tenant VMs with optional GPU. Supports larger foundation models.
Self-Hosted - Run on your own hardware. See the installation guide.
Bring Your Own Cloud - Self-host on AWS, Azure, or GCP for enterprise compliance.

Quick Start

Install the inference SDK:

pip install inference-sdk

To self-host, start a local Inference server:

pip install inference-cli && inference server start --port 9001

Then run inference:

from inference_sdk import InferenceHTTPClient

client = InferenceHTTPClient(
    # api_url="https://serverless.roboflow.com", # for Roboflow hosted inference
    api_url="http://localhost:9001",  # for self-hosted inference
    api_key="YOUR_API_KEY", # For private/fine-tuned models
)

image_url = "https://media.roboflow.com/inference/people-walking.jpg"
results = client.infer(image_url, model_id="rfdetr-small")

For more information, see Run a Model.

Roboflow App - Upload data, annotate images, train and deploy models. Get your API key.
Universe - Browse and use community datasets and models. Pass any Universe model_id directly to Inference.
Supervision - Post-process results: decode predictions, plot bounding boxes, track objects, slice images for small object detection.

Support

Community Forum - Browse questions from the community and ask your own.
Contact Support - Use live chat or contact your account representative.
Weekly Live Sessions - Join Roboflow team members and users for live Q&A.

Open Source

Core functionality is open source under Apache 2.0. View on GitHub | Contribute

Models are subject to their underlying architecture licenses (details). Cloud-connected features require a Roboflow API key.