# New Release: Ultralytics v8.4.0

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-0/1747>\
**Category:** Discussion\
**Tags:** ultralytics-official, releases, announcements\
**Created:** [January 14, 2026, 1:17pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-0/1747 "2026-01-14T13:17:52Z")\
**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:** [January 14, 2026, 1:17pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-0/1747/1 "2026-01-14T13:17:52Z")

</div>

# 🎉 Ultralytics v8.4.0 is live — introducing **Ultralytics YOLO26** 🚀

**Quick overview:** v8.4.0 delivers the **Ultralytics YOLO26** release — a re-engineered, edge-first family with **native end-to-end (NMS-free) inference**, major **CPU speedups** , and upgrades across **detect, segment, classify, pose, and OBB**. This release also includes an ARM64 Docker fix to improve deployment experience on modern hardware. 🧩

 ![Ultralytics YOLO26 Comparison Plots](https://us1.discourse-cdn.com/flex001/uploads/ultralytics1/original/2X/8/89c8804233c5d2847b6aa84e04b6992e3e5dd7ac.png)

* * *

## 🌟 New Features (Highlights)

### 🚀 Ultralytics YOLO26: native end-to-end, NMS-free

Ultralytics YOLO26 arrives in v8.4.0 with a **streamlined architecture designed for edge and low-power environments** —built for simpler exports and faster inference.

Key highlights:

- ⚡ **NMS-free end-to-end inference** (native design, reduced latency, simpler deployment)
- 🏎 **Up to 43% faster CPU inference** (edge-first optimization)
- 🧠 **MuSGD optimizer** (hybrid optimizer inspired by modern LLM training stability ideas)
- 🔧 **No DFL** (streamlines export + boosts compatibility with low-power hardware)
- 🎯 **Task-specific improvements**
  - **Segmentation** : semantic loss + multi-scale protos
  - **Pose** : RLE for higher-precision keypoints
  - **OBB** : angle loss to improve boundary behavior

- 🌐 **YOLOE-26 open-vocabulary** : zero-shot detection using text or visual prompts

* * *

## 🔧 Improvements

- 🧩 **Edge-first philosophy** : removing bottlenecks like DFL and NMS to maximize performance on constrained devices
- 🧰 **Unified & versatile** : end-to-end support across **detection, segmentation, classification, pose, and OBB**
- 🏁 **Next-gen training stability** : MuSGD helps improve convergence behavior and training stability

* * *

## 🐛 Bug Fixes

- ✅ ARM64 Dockerfile fix for smoother builds and deploys on ARM devices

* * *

## 🔀 What’s Changed (PRs)

- Fix ARM64 Dockerfile by [@glenn-jocher](https://github.com/glenn-jocher) in [PR #23175](https://github.com/ultralytics/ultralytics/pull/23175)
- `ultralytics 8.4.0` YOLO26 Models Release by [@glenn-jocher](https://github.com/glenn-jocher) in [PR #23176](https://github.com/ultralytics/ultralytics/pull/23176)

* * *

## 📦 How to Try v8.4.0

Update in seconds:

- `pip install -U ultralytics`

Quick sanity check (Python):

```python
from ultralytics import YOLO

model = YOLO("yolo26n.pt")
model.predict(source="https://ultralytics.com/images/bus.jpg")

```

You can also browse the full release details directly in the official [v8.4.0 release page](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.0), and review the full diff in the [v8.4.0 full changelog](https://github.com/ultralytics/ultralytics/compare/v8.3.253...v8.4.0).

* * *

## 💬 Feedback Welcome

Please try v8.4.0 on your workloads (especially **CPU/edge deployments** and exports), and share:

- performance numbers (CPU/GPU, batch size, `imgsz`, export target)
- export experiences (ONNX, TensorRT, CoreML, etc.)
- any regressions or odd behaviors you spot

Your feedback helps the whole YOLO community push the ecosystem forward.