[workflow action package]
@beam/predict-detectorDetect objects with an ONNX model
Loads a DETR-style object detector exported to ONNX (RT-DETR, D-FINE, DETR and the like, as Optimum and transformers.js export them) and finds the objects of an image, as labelled boxes drawn over it, or of every frame of a video, as an annotated video and one line of boxes per frame. Classes and image size come from the model's config.json and preprocessor_config.json. ONNX Runtime on the machine's accelerator (CUDA in the CUDA image, CoreML on a Mac) or the CPU; video decoded and encoded by PyAV (FFmpeg).
Trust
ExternalPlacements
3Permissions
-Versions
6[package identity]
Details
Identity and execution metadata for the latest version.
- Package
@beam/predict-detector- Repository
- -
- Author
- beam
- Category
- ml
- API version
- workflow-actions/v2
- Latest tag
- latest -> 0.1.6
- Task mode
- single-worker
- Default placement
- external-workers
- Visibility
- Public
- Published
- Nov 12, 2026, 8:04 PM
- Updated
- 11/12/2026
[workflow reference]
Use in a workflow
Drop this into a workflow definition to reference the action.
{
"actionPackageName": "@beam/predict-detector",
"actionVersionRange": "^0.1.6",
"config": {},
"inputBindings": {}
}[developer tooling]
CLI
Publish and resolve packages from a local development environment.
beam-registry publish . --tag latest curl "https://api.lab.b3m.dev/registry/v1/resolve/%40beam/predict-detector?range=latest"