Starstruk Logo MarkStarstruk

1.2.0

roboflow/rf-detr1.2.0Jul 23, 2025by SkalskiP

AI Summary

Added three new smaller model sizes (Nano, Small, Medium) and launched a dedicated documentation website.

Key Highlights

  • Introduced Nano, Small, and Medium model sizes for object detection.
  • Added `RFDETRNano`, `RFDETRSmall`, and `RFDETRMedium` classes.
  • New `deploy_to_roboflow` function for provisioning serverless cloud APIs.
  • Launched a new dedicated documentation website with tutorials.

New Features

  • New model sizes: Nano, Small, and Medium.
  • Dedicated documentation website (rfdetr.roboflow.com).
  • Cloud deployment functionality via `deploy_to_roboflow`.

Full Release Notes

## What's new šŸ”„ 

### New model sizes

RF-DETR 1.2.0 introduces three new, state-of-the-art, model sizes for object detection:

- Nano (`RFDETRNano`)
- Small (`RFDETRSmall`)
- Medium (`RFDETRMedium`)

<img width="2369" height="989" alt="image (8)" src="https://github.com/user-attachments/assets/d99aa04e-696f-4f15-9365-7f17f3b3df58" />

With the `rfdetr` Python package, you can train and run models with these architectures.

The training API is as follows:

```python
from rfdetr import RFDETRNano

model = RFDETRNano()

model.train(
    dataset_dir=<DATASET_PATH>,
    epochs=10,
    batch_size=4,
    grad_accum_steps=4,
    lr=1e-4,
    output_dir=<OUTPUT_PATH>
)
```

Trained models can also be [deployed with Roboflow Inference](https://rfdetr.roboflow.com/learn/deploy/) with the new `deploy_to_roboflow` function. This allows you to provision a serverless cloud API for running your model, as well as deploy your model in a Roboflow Workflow or with a Roboflow Inference server:

```python
from rfdetr import RFDETRNano

x = RFDETRNano(pretrain_weights="<path/to/prtrain/weights/dir>")
x.deploy_to_roboflow(
  workspace="<your-workspace>",
  project_ids=["<your-project-id>"],
  api_key="<YOUR_API_KEY>"
)
```

https://github.com/user-attachments/assets/607f8462-0e17-4777-bdd4-9be012174e42

### New documentation

[RF-DETR now has its own documentation website.](https://rfdetr.roboflow.com/) This website has tutorials on running RF-DETR with base weights, fine-tuning RF-DETR models, and deploying RF-DETR models. You can also see auto-generated docstring documentation for the main model classes.

šŸ† Contributors

@probicheaux @isaacrob-roboflow @Matvezy @MadeWithStone @SkalskiP @capjamesg