# New Release: Ultralytics v8.4.153

**URL:** <https://community.ultralytics.com/t/new-release-ultralytics-v8-4-153/2222>\
**Category:** Discussion\
**Tags:** ultralytics-official, releases, announcements\
**Created:** [September 16, 2026, 1:09pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-153/2222 "2026-09-16T13:09:31Z")\
**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:** [September 16, 2026, 1:09pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-153/2222/1 "2026-09-16T13:09:31Z")

</div>

# 🚀 Ultralytics v8.4.153 Released

**Ultralytics v8.4.153** fixes SAM3 initialization with `compile=False`, restores reliable standalone validation, preserves dataset overrides when resuming training, improves INT8 and IMX exports, and expands Ultralytics Platform annotation and billing documentation.

Users working with **SAM3, standalone validation, resumed training, IMX export, or RKNN INT8 deployment are strongly encouraged to upgrade**. 🔄

## 🌟 Key Highlights

- 🧠 SAM3 no longer triggers unintended `torch.compile` behavior when `compile=False`.
- ✅ Semantic segmentation and classification validation are more reliable.
- 🔄 Explicit `data` overrides are preserved when resuming training.
- ⚙ INT8 behavior is clearer across ONNX, TensorRT, and RKNN exports.
- 🎯 IMX export now correctly preserves `conf=0.0`.
- 🧩 Ultralytics Platform documentation now covers batch annotation, pose smart annotation, billing, referrals, and more.

## 🛠 Bug Fixes

### SAM3 initialization

[`PR #26189`](https://github.com/ultralytics/ultralytics/pull/26189) by GitHub author [`@glenn-jocher`](https://github.com/glenn-jocher) correctly converts the SAM3 predictor compilation option:

- `False` disables compilation.
- `True` enables the default compilation mode.
- Explicit compilation mode strings remain unchanged.

This prevents unintended `torch.compile` activation and related PyTorch 2.14 Dynamo errors. The fix was validated across all six Platform SAM models using HTTP and WebSocket inference.

### Standalone semantic segmentation validation

[`PR #26176`](https://github.com/ultralytics/ultralytics/pull/26176) by GitHub author [`@cainiao33`](https://github.com/cainiao33) adds background-class metadata while constructing polygon-based semantic datasets.

This prevents background pixels from being assigned to the final real class, restoring meaningful metrics such as mIoU and pixel accuracy.

### Standalone classification validation

[`PR #26182`](https://github.com/ultralytics/ultralytics/pull/26182) by GitHub author [`@cainiao33`](https://github.com/cainiao33) filters validation samples with class IDs outside the model’s supported class range.

Instead of raising an `IndexError`, validation now skips incompatible samples and provides clear warnings about mismatched classes.

### Training resume dataset overrides

[`PR #26175`](https://github.com/ultralytics/ultralytics/pull/26175) by GitHub author [`@cainiao33`](https://github.com/cainiao33) ensures that `train(resume=True, data="...")` honors the explicitly provided dataset rather than silently reverting to the checkpoint’s original dataset.

Path-based resume workflows also avoid injecting an unintended task-default dataset.

### IMX confidence thresholds

[`PR #26183`](https://github.com/ultralytics/ultralytics/pull/26183) by GitHub author [`@cainiao33`](https://github.com/cainiao33) preserves an explicit `conf=0.0` during IMX export instead of replacing it with `0.001`.

Tests now cover both `conf=0.0` and the default `conf=0.25` behavior.

## ⚙ Export Improvements

### RKNN INT8 validation

[`PR #26184`](https://github.com/ultralytics/ultralytics/pull/26184) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) correctly limits RKNN INT8 validation to detection models, including INT8-only Rockchip targets.

The accompanying examples now use a compatible detection model and dataset.

### INT8 documentation and deployment guidance

[`PR #26174`](https://github.com/ultralytics/ultralytics/pull/26174) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) aligns documentation with current export behavior:

- Quantization-aware training checkpoints can use stored INT8 ranges for ONNX and TensorRT exports without calibration data.
- TensorRT documentation no longer states that `dynamic` is automatically enabled for INT8.
- The Platform Python requirement and Hailo hardware comparison are updated.
- Huawei Ascend and Google Coral Edge TPU are included in the Hailo comparison.

## 🧩 Ultralytics Platform Updates

### Batch annotation and billing

[`PR #26186`](https://github.com/ultralytics/ultralytics/pull/26186) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) documents dataset-wide batch annotation, including:

- Class mapping
- Progress tracking
- Stopping annotation runs
- Dataset version snapshots
- Batch annotation pricing and billing

### Pose smart annotation and Platform guidance

[`PR #26185`](https://github.com/ultralytics/ultralytics/pull/26185) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) expands Platform documentation with:

- YOLO smart annotation for pose datasets
- SAM annotation support for detection, segmentation, semantic segmentation, and OBB tasks
- Supported dataset licenses
- The referrals tab
- Endpoint uptime charges
- Python 3.11 requirements
- Updated Agents availability without the previous Early Access requirement

## 📚 Documentation Updates

- [`PR #26177`](https://github.com/ultralytics/ultralytics/pull/26177) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) updates repository banner destinations to the YOLO27 preview.
- [`PR #26188`](https://github.com/ultralytics/ultralytics/pull/26188) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) links the coming-soon notice to the [YOLO27 preview and waitlist](https://www.ultralytics.com/yolo/yolo27).
- [`PR #26170`](https://github.com/ultralytics/ultralytics/pull/26170) by GitHub author [`@raimbekovm`](https://github.com/raimbekovm) removes the unreachable retired Neptune integration page.

## 📦 Upgrade Today

Install or upgrade to the latest release with:

```bash
pip install -U ultralytics

```

Review the complete changes in the [v8.4.152 to v8.4.153 comparison](https://github.com/ultralytics/ultralytics/compare/v8.4.152...v8.4.153), or visit the [Ultralytics v8.4.153 GitHub release](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.153) for release details.

Give **v8.4.153** a try and share your results, questions, or feedback with the Ultralytics YOLO community. Happy building! 🎉