Skip to content

EdgeFirst Client

The EdgeFirst Client (edgefirst-client) is the official command-line application and library for EdgeFirst Studio. It provides programmatic and command-line access to datasets, annotations, training sessions, snapshots, and model artifacts.

One install delivers everything:

python3 -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install edgefirst-client
edgefirst-client login

The PyPI wheel bundles the edgefirst-client CLI executable and the edgefirst_client Python module. No Rust toolchain is required for end users.

Language bindings

Binding Install / docs Notes
Python pip install edgefirst-client Primary target for tutorials and API reference in this manual
Rust crates.io · docs.rs Core library; powers the CLI and all bindings
Swift / iOS APPLE.md on GitHub UniFFI bindings over the Rust core
Kotlin / Android ANDROID.md on GitHub UniFFI bindings over the Rust core

Tutorial dataset

The Python tutorials use the public Coffee Cup dataset ds-145f on SaaS. Read-only examples never modify this dataset. Write examples (tutorials 06 and 07) create ephemeral sandbox datasets in your project.

See the Coffee Cup dataset zoo for background on the classes and use cases.

Reading order

  1. Command-line interface — install, authenticate, Coffee Cup quick commands, MCAP snapshot workflow
  2. Dataset import — COCO/LVIS, EdgeFirst Dataset Format, and custom Python imports
  3. Python tutorials — seven hands-on scripts aligned with the client repository examples
  4. API reference — Python API (generated) and Rust API (docs.rs)

Included with EdgeFirst Middleware

On edge devices, edgefirst-client is installed as part of the standard EdgeFirst Middleware installation. Use edgefirst-client version to confirm connectivity to EdgeFirst Studio.