# New Release: Ultralytics v8.4.184

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-184/2275>\
**Category:** Discussion\
**Tags:** releases, announcements, ultralytics-official\
**Created:** [October 23, 2026, 1:09pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-184/2275 "2026-10-23T13:09:04Z")\
**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:** [October 23, 2026, 1:09pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-184/2275/1 "2026-10-23T13:09:04Z")

</div>

# 🚀 Ultralytics v8.4.184: More Reliable Images, Models, and Workflows

**Ultralytics YOLO v8.4.184 is here!** This reliability-focused release fixes multi-page TIFF loading and improves segmentation prompts, specialized model loading, validation, experiment tracking, and Ultralytics Platform workflows. **No new model architecture is introduced** —this update focuses on making everyday work more predictable.

## 🖼 Highlight: Safer Multi-Page TIFF Loading

Grayscale TIFF scans now load as their **first page by default** , rather than incorrectly treating pages as image channels. This helps prevent scanned documents from causing channel-count errors during training.

For multispectral workflows, `cv2.IMREAD_UNCHANGED` still stacks same-size pages as bands. See [PR #26842](https://github.com/ultralytics/ultralytics/pull/26842) by @glenn-jocher for the fix.

## 🎯 Model Loading and Segmentation Improvements

- **Reliable YOLOE and YOLO-World loading:** Checkpoint contents now determine the correct model wrapper, even with generic filenames such as `best.pt`. [PR #26793](https://github.com/ultralytics/ultralytics/pull/26793) by @raimbekovm makes specialized checkpoints easier to reuse.
- **Clearer YOLOE errors:** Changing classes on a text-fused YOLOE model now raises an explicit error instead of silently relabeling predictions or failing later. [PR #26824](https://github.com/ultralytics/ultralytics/pull/26824) by @raimbekovm adds this safeguard.
- **Faithful SAM mask refinement:** Floating-point prompt masks retain negative and fractional values. [PR #26794](https://github.com/ultralytics/ultralytics/pull/26794) by @Nikhi00718 preserves the original refinement inputs.
- **Better multipart polygons:** Reversed segments connect at their closest points through [PR #26834](https://github.com/ultralytics/ultralytics/pull/26834) by @mottopanikeiku.
- **Consistent semantic backgrounds:** Class filtering preserves the polygon semantic background class with [PR #26798](https://github.com/ultralytics/ultralytics/pull/26798) by @raimbekovm.
- **Flexible model filenames:** Model file extensions are now case-insensitive thanks to [PR #26819](https://github.com/ultralytics/ultralytics/pull/26819) by @Vaishnavi220506.

## 📊 Prediction, Data, and Validation Fixes

- **Rectangular classification inputs:** [PR #26810](https://github.com/ultralytics/ultralytics/pull/26810) by @aswanth-07 fixes classification prediction with rectangular image sizes.
- **Correct tensor layout:** [PR #26801](https://github.com/ultralytics/ultralytics/pull/26801) by @Y-T-G makes BCHW tensors contiguous before the BGR-to-RGB flip.
- **Safer 2D preprocessing:** [PR #26811](https://github.com/ultralytics/ultralytics/pull/26811) by @cainiao33 prevents a `LetterBox` crash when 2D inputs need no resizing.
- **More robust downloads:** Archive URLs containing query strings now work with dataset autodownload through [PR #26789](https://github.com/ultralytics/ultralytics/pull/26789) by @cainiao33.
- **Clear stream errors:** Empty stream manifests now produce an actionable error with [PR #26791](https://github.com/ultralytics/ultralytics/pull/26791) by @cainiao33.
- **More consistent confusion matrices:** Detections are matched in confidence order through [PR #26828](https://github.com/ultralytics/ultralytics/pull/26828) by @lu123bing.
- **Cleaner label-free validation:** [PR #26837](https://github.com/ultralytics/ultralytics/pull/26837) by @raimbekovm skips class-averaged precision-recall curves when a split has no labels.
- **Leaner metric computation:** [PR #26815](https://github.com/ultralytics/ultralytics/pull/26815) by @cainiao33 moves constant setup outside `compute_ap`.

## 🔗 Smoother Integrations and Runtime Behavior

- **Respect existing W&B runs:** Training can log into a user-started run, preserve its settings, and leave it open afterward with [PR #26821](https://github.com/ultralytics/ultralytics/pull/26821) by @raimbekovm. Training arguments are also added to that run’s configuration through [PR #26831](https://github.com/ultralytics/ultralytics/pull/26831) by @raimbekovm.
- **More reliable Platform resume:** [PR #26790](https://github.com/ultralytics/ultralytics/pull/26790) by @cainiao33 prevents transient send failures from forking duplicate models.
- **Improved device handling:** [PR #26809](https://github.com/ultralytics/ultralytics/pull/26809) by @cainiao33 fixes crashes from the bare `"intel"` device string in OpenVINO and LiteRT backends.
- **Stronger settings validation:** [PR #26792](https://github.com/ultralytics/ultralytics/pull/26792) by @raimbekovm validates every `settings.update()` form and keeps fallback configuration directories private per user.

## 🛠 CI and Documentation Updates

**More complete CPU runners** now include `rsync` through [PR #26796](https://github.com/ultralytics/ultralytics/pull/26796), OpenSSH through [PR #26797](https://github.com/ultralytics/ultralytics/pull/26797), and GUI OpenCV libraries through [PR #26829](https://github.com/ultralytics/ultralytics/pull/26829), all by @glenn-jocher. GitLab mirroring also waits through rate limits instead of failing repositories with [PR #26840](https://github.com/ultralytics/ultralytics/pull/26840) by @glenn-jocher.

Documentation improvements include:

- **Corrected IMX500 export examples:** [PR #26803](https://github.com/ultralytics/ultralytics/pull/26803) by @Ali-Tharwat uses YOLO11n in the export table, while [PR #26812](https://github.com/ultralytics/ultralytics/pull/26812) and [PR #26814](https://github.com/ultralytics/ultralytics/pull/26814) by @glenn-jocher further correct the table and model naming.
- **Clearer dataset guidance:** [PR #26813](https://github.com/ultralytics/ultralytics/pull/26813) by @glenn-jocher clarifies Platform dataset downloads and Objects365 segmentation weight names.
- **More Platform training guidance:** [PR #26826](https://github.com/ultralytics/ultralytics/pull/26826) by @JaviChulvi documents detection and OBB training on segmentation datasets.
- **Refreshed Platform and export docs:** [PR #26817](https://github.com/ultralytics/ultralytics/pull/26817) by @raimbekovm updates screenshots and aligns guidance with Platform changes and Axelera’s NMS default.
- **Full-frame classification examples:** [PR #26835](https://github.com/ultralytics/ultralytics/pull/26835) by @Nikhila-Dhevi-R documents the relevant transforms.
- **Additional documentation polish:** [PR #26820](https://github.com/ultralytics/ultralytics/pull/26820) by @glenn-jocher fixes dataset and author links, and [PR #26816](https://github.com/ultralytics/ultralytics/pull/26816) by @RizwanMunawar adds a video resource.

## 🙌 Thank You, Contributors!

Thanks to everyone who contributed fixes, documentation, and testing. A special welcome to **@Ali-Tharwat** , whose first contribution arrived in [PR #26803](https://github.com/ultralytics/ultralytics/pull/26803)!

## 🚀 Try v8.4.184

Install this release with:

```bash
pip install --upgrade ultralytics==8.4.184

```

The [v8.4.184 release page](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.184) has the release details, and the [full changelog](https://github.com/ultralytics/ultralytics/compare/v8.4.183...v8.4.184) shows every change.

**Give it a try and share your feedback below!** We’d especially like to hear how the TIFF, segmentation, validation, and integration fixes work with your workflows.