# New Release: Ultralytics v8.4.50

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-50/1975>\
**Category:** Discussion\
**Tags:** ultralytics-official, releases, announcements\
**Created:** [May 13, 2026, 1:44pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-50/1975 "2026-05-13T13:44:15Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![glenn-jocher](https://sea1.discourse-cdn.com/flex001/user_avatar/community.ultralytics.com/glenn-jocher/32/83_2.png) [@glenn-jocher](https://community.ultralytics.com/u/glenn-jocher)\
**Post date:** [May 13, 2026, 1:44pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-50/1975/1 "2026-05-13T13:44:15Z")

</div>

# Ultralytics v8.4.50 is here 🚀

**Summary:** Ultralytics v8.4.50 is a deployment-focused release centered on **new DeepX export and inference support** for running **Ultralytics YOLO** models on **DeepX NPU edge hardware**. This update also improves **tuning reliability** , fixes **mixed-precision fusion issues** , and adds clearer **RT-DETR latency tuning guidance**.

## 🌟 Highlights

### DeepX export support for edge deployment 🧠⚡

The biggest addition in v8.4.50 is **DeepX export integration** , which lets you export models directly with `format="deepx"` as part of the standard Ultralytics workflow.

With DeepX export, Ultralytics now:

- converts the model through ONNX and compiles it to DeepX `.dxnn`
- creates a complete deployment bundle including:
  - compiled model file
  - `config.json` for preprocessing and calibration
  - `metadata.yaml` for model metadata

- enables `int8=True` automatically for INT8-focused deployment
- defaults to `coco128.yaml` for calibration if no dataset is provided
- supports modern tasks including **detect, segment, classify, pose, and OBB**

This makes it much easier to move YOLO models onto low-power embedded AI systems such as smart cameras, robotics platforms, and industrial devices.

### DeepX inference backend added 🔌

DeepX models exported from Ultralytics can now be loaded back for inference through a dedicated backend, making the deployment workflow more integrated end to end.

### DeepX docs and references added 📘

This release also includes documentation updates to help users get started with the new workflow more quickly, including guides, reference pages, and export documentation improvements.

## Improvements 🛠

### More honest hyperparameter tuning results ✅

Hyperparameter tuning now reports failures more accurately:

- failed runs are reflected in status output
- misleading success reporting is removed when all iterations fail
- `best_hyperparameters.yaml` is no longer written if no valid result exists

This should make tuning output much more trustworthy for experimentation and production workflows.

### Better mixed-precision fusion stability 🎯

A dtype mismatch bug in `fuse_conv_and_bn` and `fuse_deconv_and_bn` has been fixed, improving reliability when working with `float16` or `bfloat16`.

## Docs updates 📚

### RT-DETR `eval_idx` guidance for lower latency ⏱

The RT-DETR docs now explain how to use `eval_idx` to reduce decoder layers at inference time, trading a small amount of accuracy for lower latency without retraining.

## Important notes ⚠

- DeepX export currently requires **x86-64 Linux**
- **ARM64 export is not supported**
- inference support depends on the DeepX runtime and `dx_engine`

## What changed in GitHub

- **Fix tuner reporting when all iterations fail** by [@raimbekovm](https://github.com/raimbekovm) in [PR #24477](https://github.com/ultralytics/ultralytics/pull/24477)
- **Add RT-DETR `eval_idx` documentation** by [@artest08](https://github.com/artest08) in [PR #24465](https://github.com/ultralytics/ultralytics/pull/24465)
- **Fix dtype mismatch in fusion helpers** by [@raimbekovm](https://github.com/raimbekovm) in [PR #24480](https://github.com/ultralytics/ultralytics/pull/24480)
- **Add DeepX export integration** by [@ambitious-octopus](https://github.com/ambitious-octopus) in [PR #23553](https://github.com/ultralytics/ultralytics/pull/23553)

You can also explore the [full changelog for v8.4.50](https://github.com/ultralytics/ultralytics/compare/v8.4.49...v8.4.50) or view the [official v8.4.50 release](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.50).

## Try it out 🙌

If you’re targeting edge hardware, this is a great release to test. For example:

```python
from ultralytics import YOLO

model = YOLO("yolo11n.pt")
model.export(format="deepx")

```

If you’re starting a new project, we recommend using [Ultralytics YOLO26 on Ultralytics Platform](https://platform.ultralytics.com/ultralytics/yolo26) for the latest models and workflows, while [YOLO11](https://platform.ultralytics.com/ultralytics/yolo11) remains fully supported.

Please give v8.4.50 a try and let us know how it works for your deployment, training, or optimization workflows. Feedback, issues, and benchmarks are always welcome 💙