Open sourceActive

Roboflow Inference

Open-source inference engine for running Roboflow models locally via Docker server with HTTP or Python SDK access.

Open source page ↗

Field note

What it does

Free, open-source engine behind Roboflow's hosted models. Run models on your own hardware with a Docker inference server. Supports CPU, NVIDIA GPU, and Jetson. Access via HTTP API or Python Inference SDK.

Capabilities

Available capabilities

GPU and Jetson SupportVideo Processing via WebRTCHTTP API AccessLocal InferenceDocker Deployment

Tags

Tags

On-PremiseInferenceComputer Vision

Ways to use it

Ways to use it

sdk

python / inference-sdk

cli

python / inference-cli

docker

Not filed / roboflow/inference

Product features

Product features

Detect everyday objects with a ready-made model

Find objects by naming them

Get your API Key

Install the dependencies

Run the model

Run your own or another model

Use a Model API Endpoint

Where to go next