Starstruk Logo MarkStarstruk

v0.58.0

roboflow/inferencev0.58.0Oct 2, 2025by PawelPeczek-Roboflow

AI Summary

Major release introducing RF-DETR Seg (Preview) for real-time instance segmentation, alongside performance improvements for mask processing utilities.

Key Highlights

  • Added RF-DETR Seg (Preview) for state-of-the-art real-time instance segmentation.
  • Improved speed and memory usage of masks2poly and masks2multipoly.
  • Provided example code for using RF-DETR Seg with inference.

New Features

  • RF-DETR Segmentation support
  • Mask processing optimization
  • Real-time segmentation capabilities

Full Release Notes

# 💪 Added

## 🤯  RFDetr Seg — Fast, Accurate Segmentation in Roboflow 💥 
<div align="center">
<img width="65%" alt="image" src="https://github.com/user-attachments/assets/12338270-1de3-45e6-a800-aabe9d091657" />
</div>
We’re excited to announce the addition of RF-DETR Seg (Preview), bringing real-time instance segmentation to the Roboflow platform. Thanks to the outstanding work of the RF research team, users can now:

* Train and deploy RF-DETR Seg models directly in Roboflow
* Enjoy state-of-the-art accuracy with latency
* Run models seamlessly through Roboflow `inference` or the [rf-detr](https://github.com/roboflow/rf-detr) packages

This release pushes the boundary of fast, accurate segmentation — and it’s now available to explore in Roboflow. Check out our [📖  blog](https://blog.roboflow.com/rf-detr-segmentation-preview/) for more details and try out your RFDetr model in `inference`:

```python
import os
import supervision as sv
from inference import get_model
from PIL import Image
from io import BytesIO
import requests

url = "https://media.roboflow.com/dog.jpeg"
image = Image.open(BytesIO(requests.get(url).content))

model = get_model("<your-rf-model-id>")
predictions = model.infer(image)[0]
detections = sv.Detections.from_inference(predictions)
```


# 🔌 Improved
* improve speed and memory usage of masks2poly and masks2multipoly by @hansent in https://github.com/roboflow/inference/pull/1586


**Full Changelog**: https://github.com/roboflow/inference/compare/v0.57.4...v0.58.0