# New Release: Ultralytics v8.4.19

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-19/1836>\
**Category:** Discussion\
**Tags:** ultralytics-official, releases, announcements\
**Created:** [February 28, 2026, 1:16pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-19/1836 "2026-02-28T13:16:08Z")\
**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:** [February 28, 2026, 1:16pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-19/1836/1 "2026-02-28T13:16:08Z")

</div>

# 🚀 Ultralytics v8.4.19 is out!

**TL;DR:** Ultralytics `v8.4.19` makes **Ultralytics Platform training sync much more reliable** (especially model tracking via `model_id`), plus improvements for **SAM outputs** , **lightweight model stability** , and **clearer YOLO26 end-to-end docs**. ✅

If you’ve ever had a run where artifacts/checkpoints didn’t feel perfectly “attached” to the right Platform session—this release is for you. 🎯

* * *

## 🌟 New Features / Highlights

### 🔴 Platform training `model_id` fix (most important)

Training callbacks now **consistently pass `model_id`** during uploads and training events, improving continuity, cancellation handling, and final result linking.

- PR: [Platform training pass model ID fix](https://github.com/ultralytics/ultralytics/pull/23761) by [@glenn-jocher](https://github.com/glenn-jocher)
- What changed:
  - Refactored callback state into a single `trainer.platform` context (instead of scattered trainer fields), including:
    - `model_id`
    - last upload time
    - cancellation status
    - console/system loggers

  - Tightened upload + event flow for better session continuity ✅
  - Improved Platform URL handling when servers adjust model slugs (e.g., name auto-increment)

* * *

## 🧠 Improvements

### 🎯 SAM prediction cleanup (fewer duplicate overlaps)

Added NMS in `SAM3SemanticPredictor` postprocessing and feature-inference paths, reducing overlapping duplicates and keeping boxes/masks aligned.

- PR: [Add NMS to `SAM3SemanticPredictor`](https://github.com/ultralytics/ultralytics/pull/23751) by [@Laughing-q](https://github.com/Laughing-q)

### 🧱 PSA attention edge-case fix (small/lightweight configs)

Prevents zero attention heads in `PSABlock` for very small channel configs, improving robustness for lightweight/custom scales.

- PR: [Fix division by zero in PSABlock `num_heads`](https://github.com/ultralytics/ultralytics/pull/23758) by [@Y-T-G](https://github.com/Y-T-G)

### 📘 Clearer end-to-end docs for YOLO26 / YOLOv10

Clarifies that in end-to-end mode (YOLO26, YOLOv10), `max_det` and `agnostic_nms` are supported directly, and that `end2end=False` is mainly for traditional IoU-based NMS behavior.

- PR: [Document `max_det` and `agnostic_nms` support in end2end mode](https://github.com/ultralytics/ultralytics/pull/23720) by [@raimbekovm](https://github.com/raimbekovm)

* * *

## 🛠 Bug Fixes & Maintenance

### ⚙ CI updates (artifact actions bumps)

- PR: [Bump actions/upload-artifact from 6 to 7](https://github.com/ultralytics/ultralytics/pull/23748) by [@dependabot[bot]](https://github.com/dependabot)
- PR: [Bump actions/download-artifact from 7 to 8](https://github.com/ultralytics/ultralytics/pull/23749) by [@dependabot[bot]](https://github.com/dependabot)

### 🎥 Docs media update

Updated the African Wildlife dataset tutorial embed.

- PR: [Add tutorial video to docs](https://github.com/ultralytics/ultralytics/pull/23757) by [@RizwanMunawar](https://github.com/RizwanMunawar)

* * *

## 🧪 Try it now

Update:

```bash
pip install -U ultralytics

```

Quick sanity check (Ultralytics YOLO):

```python
from ultralytics import YOLO

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

```

If you’re using Ultralytics Platform, this release should noticeably improve training run tracking—especially around model uploads and session linking. 📡

* * *

## 🔗 Release links

Grab details from the [v8.4.19 release page](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.19) and browse the [full v8.4.18…v8.4.19 changelog](https://github.com/ultralytics/ultralytics/compare/v8.4.18...v8.4.19).

* * *

## 💬 Feedback welcome

Please try `v8.4.19` and share:

- any Platform training edge-cases you still hit 🧩
- SAM prediction quality differences you notice 🎯
- lightweight/custom model configs that benefit from the PSA fix 🧱

Thanks to everyone contributing—this pace is powered by the Ultralytics team and the wider YOLO community. 🙌