# New Release: Ultralytics v8.4.143 **URL:** \ **Category:** Discussion\ **Tags:** ultralytics-official, releases, announcements\ **Created:** [September 8, 2026, 1:10pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-143/2208 "2026-09-08T13:10:55Z")\ **Posts on this page:** 1\ **Page:** 1
**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 8, 2026, 1:10pm UTC](https://community.ultralytics.com/t/new-release-ultralytics-v8-4-143/2208/1 "2026-09-08T13:10:55Z")
# πŸš€ Ultralytics v8.4.143: INT8 QAT for YOLO26, Better Metrics, and Smoother Deployment Ultralytics **v8.4.143** is now available! This release introduces **INT8 quantization-aware training for YOLO26** , adds **per-class AP75 metrics** , improves deployment and Platform workflows, and delivers important training, lifecycle, and documentation fixes. Explore the [v8.4.143 GitHub release](https://github.com/ultralytics/ultralytics/releases/tag/v8.4.143) for the complete release details. πŸŽ‰ ## 🌟 New Features ### 🧠 INT8 Quantization-Aware Training for YOLO26 You can now use `quantize=8` during training to fine-tune pretrained YOLO26 models while simulating INT8 quantization: ```bash yolo train model=yolo26n.pt data=coco8.yaml quantize=8 ``` The resulting checkpoint stores calibrated quantization ranges and can be exported directly to ONNX or TensorRT with Q/DQ nodesβ€”without needing calibration data during export. This can help INT8 deployments retain more accuracy while reducing memory use, inference latency, and power consumption on resource-constrained edge hardware. Implemented in [PR #26083](https://github.com/ultralytics/ultralytics/pull/26083) by [@Bovey0809](https://github.com/Bovey0809). ### πŸ“Š Per-Class AP75 Metrics Validation results now include per-class AP75 values in: - Validation summaries - DataFrame outputs - CSV files - JSON exports The update applies consistently across detection, segmentation, pose, and OBB tasks, making category-level model analysis more informative. Added in [PR #24489](https://github.com/ultralytics/ultralytics/pull/24489) by [@woreom](https://github.com/woreom). ## πŸš€ Deployment and Integration Improvements - YOLO26 DeepStream guidance now uses the native Ultralytics exporter instead of a separate third-party script through [PR #26077](https://github.com/ultralytics/ultralytics/pull/26077) by [@raimbekovm](https://github.com/raimbekovm). - Examples now explicitly select the NMS-free YOLO26 head with `nms=False` where required, including Triton, DALI, SAM, and C++ workflows, through [PR #26072](https://github.com/ultralytics/ultralytics/pull/26072) by [@raimbekovm](https://github.com/raimbekovm). - Quantization guidance no longer promises FP32 output for export formats that do not support it, following [PR #26080](https://github.com/ultralytics/ultralytics/pull/26080) by [@raimbekovm](https://github.com/raimbekovm). - Documentation now covers Apple Core AI export, spatial Albumentations support, and updated `max_det` dataset behavior through [PR #26078](https://github.com/ultralytics/ultralytics/pull/26078) by [@raimbekovm](https://github.com/raimbekovm). Together, these changes simplify deployment across DeepStream, TensorRT, Core AI, OpenVINO, Edge TPU, Jetson, and other supported environments. ⚑ ## πŸ›  Training and Reliability Fixes - Pretrained-model state is now handled correctly after `reset_weights()` and in `MultiTrainer`, reducing the risk of unintentionally training from random initialization. See [PR #26084](https://github.com/ultralytics/ultralytics/pull/26084) by [@raimbekovm](https://github.com/raimbekovm). - Copies and conversions of `Results` objects now preserve prediction save directories through [PR #26074](https://github.com/ultralytics/ultralytics/pull/26074) by [@aswanth-07](https://github.com/aswanth-07). - Stale SAM3 decoder coordinate caches are removed to improve reliability across device and precision changes in [PR #26075](https://github.com/ultralytics/ultralytics/pull/26075) by [@tritsystem](https://github.com/tritsystem). ## 🌍 Platform and Authentication Updates Ultralytics now uses the Platform SDK for shared login, model downloads, dataset exports, and access to training resources. - Platform SDK integration arrives in [PR #26087](https://github.com/ultralytics/ultralytics/pull/26087) by [@glenn-jocher](https://github.com/glenn-jocher). - Python 3.11 and newer environments now require Platform SDK `>=0.1.32` for shared login through [PR #26088](https://github.com/ultralytics/ultralytics/pull/26088) by [@glenn-jocher](https://github.com/glenn-jocher). Local workflows on older supported Python versions remain available. ## πŸ“š Documentation Refresh A broad documentation review improves correctness and consistency across: - CLI syntax and Python examples - Export formats and precision guidance - Dataset layouts and image decoding - Tracker behavior - Hardware and deployment integrations - YOLO26 depth-estimation dataset guidance - Albumentations support for boxes, polygons, keypoints, depth maps, and semantic masks - Ultralytics Platform dataset and training links These updates are included in [PR #26070](https://github.com/ultralytics/ultralytics/pull/26070) by [@glenn-jocher](https://github.com/glenn-jocher). ## πŸ‘‹ Welcome, New Contributors! A warm welcome and thank you to: - [@tritsystem](https://github.com/tritsystem), who made their first contribution in [PR #26075](https://github.com/ultralytics/ultralytics/pull/26075). - [@woreom](https://github.com/woreom), who made their first contribution in [PR #24489](https://github.com/ultralytics/ultralytics/pull/24489). ## πŸ“¦ Try v8.4.143 Upgrade to the latest release with: ```bash pip install -U ultralytics ``` Please try the new QAT workflow, validation metrics, and deployment improvements, then share your feedback and any issues with the community. You can review every change in the [full v8.4.142-to-v8.4.143 changelog](https://github.com/ultralytics/ultralytics/compare/v8.4.142...v8.4.143). πŸ™Œ