# New Release: Ultralytics v8.4.54

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-54/1987>\
**Category:** Discussion\
**Tags:** releases, announcements, ultralytics-official\
**Created:** [May 26, 2026, 2:02pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-54/1987 "2026-05-26T14:02:55Z")\
**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 26, 2026, 2:02pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-54/1987/1 "2026-05-26T14:02:55Z")

</div>

# Ultralytics v8.4.54 is here 🚀

**Quick summary:** Ultralytics **v8.4.54** brings a major deployment upgrade with **local Qualcomm QNN export for Snapdragon devices** 📱⚡, plus cleaner export docs, `imgsz` support across all Solutions, improved tuning compatibility, and several reliability fixes for segmentation and image caching.

If you’re building with **Ultralytics YOLO** , especially for edge and mobile deployment, this is a great release to try.

## 🌟 Highlights

### 📱 New: Qualcomm QNN export for Snapdragon devices

The headline feature in **v8.4.54** is a new `qnn` export format, making it much easier to deploy YOLO models on Qualcomm Snapdragon hardware.

With `format="qnn"`, you can now export a model **locally** into a QNN-ready package using ONNX Runtime’s QNN Execution Provider:

```python
from ultralytics import YOLO

model = YOLO("yolo26n.pt")
model.export(format="qnn")

```

Why this matters:

- ✅ No Qualcomm account required
- ✅ No cloud upload required
- ✅ No separate Qualcomm SDK download required
- ✅ Supports Snapdragon deployment paths including **CPU, GPU, and Hexagon NPU**

This is a big step forward for mobile, embedded, automotive, and low-power AI workflows.

### 🔌 Better QNN workflow integration

QNN is now integrated more cleanly into the standard Ultralytics export and backend flow:

- Registered as a supported export target
- Dedicated backend support for inference and validation workflows
- Automatic disabling of end-to-end export where QNN does not support it, helping avoid invalid configurations

### 📘 Export docs cleanup and standardization

Deployment documentation has been reorganized into a clearer **Export / Predict / Validate** structure across formats.

This release also adds a dedicated guide for Qualcomm QNN and improves the overall deployment experience across formats such as ONNX, TensorRT, OpenVINO, NCNN, TFLite, and CoreML. You can explore the latest docs in the [Ultralytics documentation](https://docs.ultralytics.com/).

## Improvements

### 🖼 `imgsz` support added to all Solutions

All Solutions now support the `imgsz` argument, giving users more direct control over inference input size and the speed/accuracy tradeoff. Logging was also improved so reported input shapes better match actual model input.

### 🎯 Ray Tune compatibility update

Tuner trial ID handling was updated for compatibility with **Ray 2.7+** , while keeping support for older versions.

### 👀 Docs updates for validation and prediction

A couple of nice usability improvements landed in the docs:

- Validation docs now include `show_labels` and `show_conf`
- Prediction docs now explicitly list `.jpeg2000` as a supported image format

## Bug Fixes

### 🛠 Segmentation and data-loading reliability

This release includes a few important fixes to make training and inference more robust:

- Fixed semantic segmentation mask loading when masks appear as `(H, W, 1)`
- Fixed stale `.npy` cache issues that could cause grayscale models to load RGB data unexpectedly
- Improved dtype consistency in semantic segmentation loss, especially helpful for mixed-precision training

### 🧠 Tracking utility optimization

Some SciPy usage in tracking internals was replaced with NumPy equivalents, helping simplify dependencies and improve maintainability.

## What changed

- Docs: add `show_labels` and `show_conf` val args by [@K-saif](https://github.com/K-saif) in [PR #24571](https://github.com/ultralytics/ultralytics/pull/24571)
- Scope scipy imports and optimize `embedding_distance` with NumPy by [@Laughing-q](https://github.com/Laughing-q) in [PR #24572](https://github.com/ultralytics/ultralytics/pull/24572)
- Revert `_gaussian_smooth` changes by [@Laughing-q](https://github.com/Laughing-q) in [PR #24575](https://github.com/ultralytics/ultralytics/pull/24575)
- Match `aux_loss` dtype to `ce_loss` in `SemanticSegmentationLoss` by [@ahmet-f-gumustas](https://github.com/ahmet-f-gumustas) in [PR #24576](https://github.com/ultralytics/ultralytics/pull/24576)
- Add `imgsz` support to all Ultralytics Solutions by [@Laughing-q](https://github.com/Laughing-q) in [PR #24579](https://github.com/ultralytics/ultralytics/pull/24579)
- Fix: squeeze single-channel 3D mask arrays from patched `cv2.imread` by [@Laughing-q](https://github.com/Laughing-q) in [PR #24582](https://github.com/ultralytics/ultralytics/pull/24582)
- Fix grayscale test failure from stale `.npy` cache files by [@Laughing-q](https://github.com/Laughing-q) in [PR #24583](https://github.com/ultralytics/ultralytics/pull/24583)
- Add jpeg2000 to predict images formats table by [@raimbekovm](https://github.com/raimbekovm) in [PR #24588](https://github.com/ultralytics/ultralytics/pull/24588)
- Fix tuner trial ID retrieval for Ray 2.7+ compatibility by [@lmycross](https://github.com/lmycross) in [PR #24590](https://github.com/ultralytics/ultralytics/pull/24590)
- `ultralytics 8.4.54` Qualcomm QNN export by [@glenn-jocher](https://github.com/glenn-jocher) in [PR #24591](https://github.com/ultralytics/ultralytics/pull/24591)

## Why upgrade?

Ultralytics **v8.4.54** is especially valuable if you:

- Deploy on **Qualcomm Snapdragon**
- Want a smoother path from training to edge inference
- Use Solutions and need configurable `imgsz`
- Run segmentation pipelines with mixed or unusual mask formats
- Depend on Ray Tune in newer environments

For new projects, we recommend starting with [Ultralytics YOLO26](https://platform.ultralytics.com/ultralytics/yolo26), our latest stable model family. If you’re managing datasets, training, and deployment workflows, the [Ultralytics Platform](https://platform.ultralytics.com/) remains the simplest way to get started.

## Try it out 🙌

You can upgrade with:

```bash
pip install -U ultralytics

```

See the [v8.4.54 release page](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.54) for the full release, and browse the [full changelog](https://github.com/ultralytics/ultralytics/compare/v8.4.53...v8.4.54) for all details.

We’d love your feedback — especially if you test the new QNN export flow on Snapdragon devices.