# New Release: Ultralytics v8.4.183

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-183/2273>\
**Category:** Discussion\
**Tags:** announcements, ultralytics-official, releases\
**Created:** [October 20, 2026, 1:10pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-183/2273 "2026-10-20T13:10:40Z")\
**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 20, 2026, 1:10pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-183/2273/1 "2026-10-20T13:10:40Z")

</div>

# 🚀 Ultralytics v8.4.183: Broader Hardware Support, More Reliable Workflows

**Ultralytics YOLO v8.4.183 expands YOLO26 deployment on Axelera and Intel hardware, boosts batched inference throughput, and improves reliability across training, prediction, and evaluation.** This release also brings fixes for dataset handling, exports, and Ultralytics Platform workflows.

## ⚡ Hardware and Deployment Highlights

- **Every YOLO26 task on Axelera Metis:** [PR #26782 by @a-tangfan](https://github.com/ultralytics/ultralytics/pull/26782) adds Voyager SDK 1.9.0 support for detection, instance segmentation, semantic segmentation, depth estimation, pose, OBB, and classification through export, prediction, and validation. Detection uses the NMS-free path by default.
  - The documented COCO128 benchmark reaches **109 images/second at batch 32** , compared with 40 images/second on SDK 1.8. Results depend on hardware and workload; batch 1 remains useful for latency-sensitive applications.
  - Existing SDK 1.8 exports remain compatible. The updated setup requires `metis-dkms` 1.7.0 or newer and PyTorch below 2.15.

- **Intel acceleration for LiteRT:** [PR #26779 by @onuralpszr](https://github.com/ultralytics/ultralytics/pull/26779) adds Intel CPU and NPU inference support, including YOLO26 task types.

## 🏋 Training and Resume Improvements

- **More reliable warmup and gradient accumulation:** [PR #26753 by @cainiao33](https://github.com/ultralytics/ultralytics/pull/26753) clamps the warmup accumulation ramp to its capped nominal value.
- **Reduced MPS memory growth:** [PR #26744 by @vtarasv](https://github.com/ultralytics/ultralytics/pull/26744) buckets target padding to prevent per-shape driver memory growth during training.
- **Consistent Platform resumes:** [PR #26689 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26689) keeps resumed local training under the same Platform model.
- **More robust checkpoint and logging resumes:** [PR #26777 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26777) fixes distillation resumes under restricted checkpoint loading, while [PR #26772 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26772) restores W&B resumes on Windows without the `latest-run` symlink.
- **Better depth calibration selection:** [PR #26778 by @Y-T-G](https://github.com/ultralytics/ultralytics/pull/26778) preserves an existing calibration when it scores best.

## 📊 Data and Evaluation Fixes

- **Correct ImageNet-subset validation:** [PR #26760 by @Nikhi00718](https://github.com/ultralytics/ultralytics/pull/26760) aligns subset folders with classifier class IDs. In the reported ImageNette test, this corrects validation accuracy from **9.45% to 70.37%** —an evaluation fix, not a model-training gain.
- **Improved UTF-8 handling:** [PR #26745 by @MohammadHijjawi97](https://github.com/ultralytics/ultralytics/pull/26745) preserves non-ASCII grounding annotations, and [PR #26761 by @mottopanikeiku](https://github.com/ultralytics/ultralytics/pull/26761) accepts labels containing a UTF-8 byte-order mark.
- **COCO RLE segmentation conversion:** [PR #26774 by @Vaishnavi220506](https://github.com/ultralytics/ultralytics/pull/26774) adds mask decoding in `convert_coco`.
- **Correct YOLOE validation setup:** [PR #26770 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26770) converts NDJSON validation data before prompt setup.
- **More flexible source handling:** [PR #26781 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26781) makes URL-scheme and archive-extension comparisons case-insensitive.

## 🛠 Prediction, Export, and Visualization Fixes

- **Consistent image inputs:** [PR #26741 by @Nikhi00718](https://github.com/ultralytics/ultralytics/pull/26741) scales 16-bit NumPy inputs to 8 bits, matching image-file prediction behavior.
- **Reliable FastSAM exclusions:** [PR #26759 by @Nikhi00718](https://github.com/ultralytics/ultralytics/pull/26759) ensures negative points exclude masks regardless of prompt order.
- **Safer INT8 calibration:** [PR #26750 by @Y-T-G](https://github.com/ultralytics/ultralytics/pull/26750) fixes crashes with `rect=True`, while [PR #26783 by @Nikhi00718](https://github.com/ultralytics/ultralytics/pull/26783) preserves rectangular classification calibration shapes.
- **Grayscale LiteRT compatibility:** [PR #26747 by @aswanth-07](https://github.com/ultralytics/ultralytics/pull/26747) corrects layout detection for grayscale exports.
- **Thin-image and depth-map handling:** [PR #26771 by @aswanth-07](https://github.com/ultralytics/ultralytics/pull/26771) keeps letterbox dimensions positive, and [PR #26754 by @cainiao33](https://github.com/ultralytics/ultralytics/pull/26754) preserves two-dimensional depth maps for single-row, single-column, and 1 × 1 outputs.
- **Correct semantic overlay colors:** [PR #26755 by @cainiao33](https://github.com/ultralytics/ultralytics/pull/26755) fixes swapped red and blue channels in training and validation mosaics.
- **Additional reliability fixes:** [PR #26766 by @cainiao33](https://github.com/ultralytics/ultralytics/pull/26766) addresses warmup endpoints, `Conv2` fusion, BOM source lists, image-opening errors, semantic mask names, and float64 OBB corners.
- **Safer temporary imports:** [PR #26756 by @Vaishnavi220506](https://github.com/ultralytics/ultralytics/pull/26756) restores pre-existing temporary module aliases.

## 📚 Platform, Documentation, and Maintenance

- [PR #26765 by @glenn-jocher](https://github.com/ultralytics/ultralytics/pull/26765) documents Gemma 4 and Nemotron as hosted Platform annotation models.
- [PR #26775 by @raimbekovm](https://github.com/ultralytics/ultralytics/pull/26775) clarifies Platform visibility, Annotate similar, TensorRT JetPack 6 troubleshooting, and DOTA submission flags.
- [PR #26785 by @glenn-jocher](https://github.com/ultralytics/ultralytics/pull/26785) requires `ultralytics-thop>=2.2.3` for functional linear and attention operation counts.
- [PR #26776 by @UltralyticsAssistant](https://github.com/ultralytics/ultralytics/pull/26776) updates the self-hosted runner cleanup action.
- [PR #26788 by @glenn-jocher](https://github.com/ultralytics/ultralytics/pull/26788) packages the v8.4.183 release.

## 🙌 Thank You, Contributors!

Thanks to the YOLO community and Ultralytics team for these improvements. A special welcome to first-time contributors **@vtarasv** and **@mottopanikeiku**!

## 🚀 Try It and Share Your Feedback

Upgrade to the announced version:

```bash
pip install --upgrade ultralytics==8.4.183

```

Explore the [v8.4.183 release notes](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.183) for details, or review the [full changelog](https://github.com/ultralytics/ultralytics/compare/v8.4.182...v8.4.183) to see everything included.

**How does v8.4.183 perform in your workflow?** Share your hardware, task, benchmarks, and any issues below—your feedback helps improve future releases.