# New Release: Ultralytics v8.3.229

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-3-229/1652>\
**Category:** Discussion\
**Tags:** ultralytics-official, releases, announcements\
**Created:** [November 19, 2025, 1:14pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-3-229/1652 "2025-11-19T13:14:22Z")\
**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:** [November 19, 2025, 1:14pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-3-229/1652/1 "2025-11-19T13:14:22Z")

</div>

# 🚀 Ultralytics v8.3.229 Release – 3× Faster COCO Segmentation, Safer Environments, Better Docs

Ultralytics v8.3.229 is now live! This release focuses on **much faster COCO instance segmentation validation (~3×)**, **safer dependency and environment controls** , **more reliable multi‑GPU behavior** , and a **big documentation refresh** around real‑world deployments (Jetson, datasets, integrations).

If you’re training or validating **YOLO11-seg** on COCO‑style datasets or running YOLO in production environments, this one is for you. ⚡🧠

* * *

## 🌟 Summary

- ⚡ **Up to 300% faster COCO instance segmentation validation** via a new in‑repo RLE pipeline.
- 🛡 **Tighter dependency bounds and environment checks** for more predictable installs and exports.
- 🧮 **More robust multi‑GPU training & validation** , especially on small datasets.
- 📚 **Updated docs and guides** for Jetson AGX Thor (JetPack 7), datasets, and deployment stacks.
- 💬 **Improved docs build & chat widget** experience for easier in‑docs assistance.

You can review the full release notes in the [v8.3.229 GitHub release page](https://github.com/ultralytics/ultralytics/releases/tag/v8.3.229).

* * *

## ✨ New Features & Performance

### ⚡ 300% Faster COCO Segmentation Validation

Thanks to a major optimization by [@Laughing-q](https://github.com/Laughing-q) in [PR #22651](https://github.com/ultralytics/ultralytics/pull/22651), COCO instance segmentation validation is now **up to 3× faster** :

- New **native, optimized RLE encoding** pipeline for masks.
- Cleaner, more efficient mask scaling using `ops.scale_masks` with `ratio_pad` support.
- Removal of per‑mask thread pools for simpler and more predictable behavior.

This especially benefits frequent validation runs for **YOLO11 segmentation models** on COCO‑like datasets. 🏎

* * *

## 🔐 Environment & Dependency Improvements

We’ve put a lot of work into making installs and exports more stable and reproducible:

- [@onuralpszr](https://github.com/onuralpszr) added **upper bounds on core dependencies** like `numpy`, `torch`, `opencv`, `pillow`, `polars`, etc. in [PR #22701](https://github.com/ultralytics/ultralytics/pull/22701) to avoid surprise breakages from upstream updates.
- [@Y-T-G](https://github.com/Y-T-G) enforced **IMX‑specific export guards** in [PR #22673](https://github.com/ultralytics/ultralytics/pull/22673), explicitly asserting `python>=3.8<3.12` and `torch<2.9` for Sony IMX exports with clear error messages.
- The official Docker image now uses **PyTorch 2.9.1, CUDA 12.8, cuDNN 9** via [PR #22708](https://github.com/ultralytics/ultralytics/pull/22708) by [@glenn-jocher](https://github.com/glenn-jocher).
- Docker builds are simpler and more robust after removing the explicit `pip install uv` in favor of the base image tools, via [PR #22709](https://github.com/ultralytics/ultralytics/pull/22709) by [@glenn-jocher](https://github.com/glenn-jocher).

These changes should significantly reduce “it worked yesterday, broken today” issues when you reinstall or update environments. 🛡

* * *

## 🧮 Multi‑GPU & CI Reliability

Multi‑GPU workflows and CI coverage were strengthened to catch edge cases earlier:

- [@Y-T-G](https://github.com/Y-T-G) fixed a **distributed sampler edge case** where `batch_size` could be larger than the dataset size, now safely falling back to `batch_size=1` in [PR #22714](https://github.com/ultralytics/ultralytics/pull/22714).
- GPU CI is **re‑enabled with modern hardware** (RTX6000 Pro `gpu-latest` runners) and more realistic training tests like `coco128`, thanks to [PR #22710](https://github.com/ultralytics/ultralytics/pull/22710) by [@glenn-jocher](https://github.com/glenn-jocher).
- Clearer **GPU auto‑selection logs and warnings** when fewer idle GPUs than requested are available were added in [PR #22716](https://github.com/ultralytics/ultralytics/pull/22716) by [@glenn-jocher](https://github.com/glenn-jocher).
- GitHub Actions runner images were bumped to `v2.329.0` via [PR #22707](https://github.com/ultralytics/ultralytics/pull/22707) by [@glenn-jocher](https://github.com/glenn-jocher) for a more modern CI base.

You should see more predictable behavior when using multiple GPUs, especially on small custom datasets.

* * *

## 📚 Documentation & Guides

This release includes a wide‑ranging docs refresh focused on **modern hardware and practical deployments** :

- A large documentation review by [@glenn-jocher](https://github.com/glenn-jocher) in [PR #22698](https://github.com/ultralytics/ultralytics/pull/22698) brings:
  - Consistent **YOLO11 branding** and updated licensing text.
  - Updated setup and deployment guides for Docker, AzureML, Raspberry Pi, DeepStream, Coral Edge TPU, Vertex AI, and more.
  - Refreshed dataset pages (COCO variants, MNIST, DOTA, pose & segment subsets, tracking, Explorer) with clearer structures and examples.

- [@lakshanthad](https://github.com/lakshanthad) updated the **NVIDIA Jetson guide** for **Jetson AGX Thor (JetPack 7)** including specs, flashing steps, and new benchmarks in [PR #22699](https://github.com/ultralytics/ultralytics/pull/22699).
- [@RizwanMunawar](https://github.com/RizwanMunawar) embedded a **step‑by‑step YOLO11 training video** on the KITTI dataset page via [PR #22685](https://github.com/ultralytics/ultralytics/pull/22685) to help new users get started visually.

All of this aims to make it easier to move from “quick experiment” to **production‑ready YOLO** on real devices.

* * *

## 🧩 Docs Build, UI & Chat Enhancements

The docs experience and LLM chat helper were improved in several steps:

- HTML minification was made more efficient and **safer** to avoid breaking scripts or code blocks in [PR #22676](https://github.com/ultralytics/ultralytics/pull/22676) and [PR #22678](https://github.com/ultralytics/ultralytics/pull/22678) by [@glenn-jocher](https://github.com/glenn-jocher), preserving `script`, `pre`, `code`, and `textarea` content and improving cleanup.
- The local docs server now has clearer logging and better error handling for a smoother local docs dev workflow, also in [PR #22676](https://github.com/ultralytics/ultralytics/pull/22676).
- The in‑docs chat widget was iteratively upgraded:
  - New `chatExamples` argument and support for [Ultralytics LLM](https://github.com/ultralytics/ultralytics/pull/22680) in [PR #22680](https://github.com/ultralytics/ultralytics/pull/22680) by [@glenn-jocher](https://github.com/glenn-jocher).
  - An update to `ultralytics/llm@v0.0.5` in [PR #22691](https://github.com/ultralytics/ultralytics/pull/22691) by [@glenn-jocher](https://github.com/glenn-jocher).
  - A bump to `chat.js v0.0.6` in [PR #22696](https://github.com/ultralytics/ultralytics/pull/22696) by [@glenn-jocher](https://github.com/glenn-jocher) with separate `chatExamples` and `searchExamples` for better targeted help.

- Several fixes around extra JS, Explorer page 404s, and quoting bugs in the chat were addressed in:
  - [PR #22663](https://github.com/ultralytics/ultralytics/pull/22663) by [@glenn-jocher](https://github.com/glenn-jocher)
  - [PR #22666](https://github.com/ultralytics/ultralytics/pull/22666) by [@glenn-jocher](https://github.com/glenn-jocher)

The result is a **faster, more robust docs site** with richer, context‑aware assistance.

* * *

## 🎨 Repo & Config Quality

Code style and config consistency were also improved:

- Prettier’s default `print-width` was updated to 120 and applied across JS/HTML/CSS and dataset YAML files in [PR #22706](https://github.com/ultralytics/ultralytics/pull/22706) by [@glenn-jocher](https://github.com/glenn-jocher), resulting in cleaner diffs and improved readability.
- Many docs tables and argument macros for `train`, `val`, `export`, `predict`, solutions, and visualization were tightened up to better match current APIs and best practices.

* * *

## 👥 PR Highlights & Contributors

A big thank you to everyone who contributed to this release (in alphabetical order of PR title):

- [@glenn-jocher](https://github.com/glenn-jocher) – [Update extra.js (PR #22663)](https://github.com/ultralytics/ultralytics/pull/22663)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Fix Explorer page 404 (PR #22666)](https://github.com/ultralytics/ultralytics/pull/22666)
- [@glenn-jocher](https://github.com/glenn-jocher) – [4% ➜ 7% HTML minify efficiency in docs build (PR #22676)](https://github.com/ultralytics/ultralytics/pull/22676)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Preserve `script` tags in JS minify (PR #22678)](https://github.com/ultralytics/ultralytics/pull/22678)
- [@glenn-jocher](https://github.com/glenn-jocher) – [New `chatExamples` arg (PR #22680)](https://github.com/ultralytics/ultralytics/pull/22680)
- [@Y-T-G](https://github.com/Y-T-G) – [Assert `python>=3.8<3.12` and `torch<2.9` for IMX export (PR #22673)](https://github.com/ultralytics/ultralytics/pull/22673)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Use `ultralytics/llm@v0.0.5` (PR #22691)](https://github.com/ultralytics/ultralytics/pull/22691)
- [@RizwanMunawar](https://github.com/RizwanMunawar) – [Add YOLO11 KITTI training video to docs (PR #22685)](https://github.com/ultralytics/ultralytics/pull/22685)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Update `mkdocs.yml` to `chat.js v0.0.6` (PR #22696)](https://github.com/ultralytics/ultralytics/pull/22696)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Documentation review and fixes (PR #22698)](https://github.com/ultralytics/ultralytics/pull/22698)
- [@lakshanthad](https://github.com/lakshanthad) – [Update docs with NVIDIA Jetson AGX Thor (JetPack 7) + benchmarks (PR #22699)](https://github.com/ultralytics/ultralytics/pull/22699)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Update Prettier default `print-width 120` (PR #22706)](https://github.com/ultralytics/ultralytics/pull/22706)
- [@onuralpszr](https://github.com/onuralpszr) – [Update dependency versions in `pyproject.toml` to enforce upper limits (PR #22701)](https://github.com/ultralytics/ultralytics/pull/22701)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Update Dockerfile to `2.9.1-cuda12.8-cudnn9-runtime` (PR #22708)](https://github.com/ultralytics/ultralytics/pull/22708)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Update Dockerfile-runner to `v2.329.0` (PR #22707)](https://github.com/ultralytics/ultralytics/pull/22707)
- [@Y-T-G](https://github.com/Y-T-G) – [Handle multi-GPU validation when `batch_size` \> dataset size (PR #22714)](https://github.com/ultralytics/ultralytics/pull/22714)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Remove `pip install uv` from Dockerfile (PR #22709)](https://github.com/ultralytics/ultralytics/pull/22709)
- [@glenn-jocher](https://github.com/glenn-jocher) – [Print warning when fewer GPUs are selected (PR #22716)](https://github.com/ultralytics/ultralytics/pull/22716)
- [@glenn-jocher](https://github.com/glenn-jocher) – [New RTX6000 Pro `gpu-latest` runners in `ci.yml` (PR #22710)](https://github.com/ultralytics/ultralytics/pull/22710)
- [@Laughing-q](https://github.com/Laughing-q) – [`ultralytics 8.3.229` 300% faster COCO Segmentation val (PR #22651)](https://github.com/ultralytics/ultralytics/pull/22651)

You can see all changes in the [full v8.3.229 changelog diff](https://github.com/ultralytics/ultralytics/compare/v8.3.228...v8.3.229).

* * *

## 🧪 How to Upgrade and Try It

Upgrade to v8.3.229 with:

```bash
pip install -U ultralytics

```

Then run a quick validation (for example on a YOLO11 segmentation model):

```bash
yolo val model=yolo11x-seg.pt data=coco.yaml split=val

```

You should notice much faster COCO segmentation validation compared to earlier versions, especially on larger val sets.

* * *

## 💬 Feedback & Discussion

We’d love to hear how v8.3.229 works for you:

- Do you see the expected speedup in COCO segmentation validation?
- Are the new environment guards and dependency bounds helping with installs and exports?
- Is the updated Jetson / dataset / deployment documentation clearer for your workflows?

Please share your benchmarks, issues, and suggestions in this Discourse thread or open an issue or discussion in the [Ultralytics GitHub repository](https://github.com/ultralytics/ultralytics).

Thanks again to the entire YOLO community and Ultralytics team for driving these improvements forward. Your real‑world feedback is what keeps YOLO fast, stable, and practical. 🚀