Ultralytics YOLO

🌟 Summary

Ultralytics 8.4.183 expands hardware deployment and improves reliability across training, prediction, data handling, and evaluation. 🚀

📊 Key Changes

• Axelera Metis support for every YOLO26 task: With Voyager SDK 1.9.0, detection, instance and semantic segmentation, depth estimation, pose, OBB, and classification models can all export, predict, and validate through yolo. The NMS-free detection path is used by default. SDK 1.8 exports remain compatible; the updated setup requires metis-dkms 1.7.0 or newer and supports PyTorch versions below 2.15.
• Higher Axelera throughput: In the documented COCO128 benchmark, SDK 1.9 reached 109 images per second at batch 32, compared with 40 images per second on SDK 1.8. Larger prediction batches can improve throughput; batch size 1 remains useful when per-image latency matters.
• Intel acceleration for LiteRT models: LiteRT exports can now run on Intel CPUs and NPUs, including support for YOLO26 task types such as detection, segmentation, depth, pose, OBB, and classification.
• More reliable training and depth calibration: Fixes address warmup and gradient accumulation edge cases, MPS memory growth during training, and depth calibration replacing an already-good calibration with a worse one. Interrupted runs can also resume under the same Ultralytics Platform model, and W&B resuming works on Windows.
• More accurate and robust data workflows: Fixes improve ImageNet-subset classifier validation, accept UTF-8 labels with a byte-order mark, preserve non-ASCII grounding annotations, convert COCO RLE masks for segmentation, and align 16-bit NumPy images with image-file prediction behavior.
• Prediction, export, and visualization fixes: Negative FastSAM points now reliably exclude masks regardless of prompt order. Other fixes cover very thin images, rectangular INT8 calibration, grayscale LiteRT models, semantic overlay colors, depth-map saving, and several additional edge cases.
• Platform and documentation updates: Platform documentation now includes Gemma 4 and Nemotron as hosted annotation models and explains new visibility controls and workflows. Benchmark instructions were also clarified for NMS-free evaluation.

🎯 Purpose & Impact

• Deploy YOLO26 on more hardware and tasks: Axelera users can use one integrated workflow across all seven YOLO26 tasks, while Intel LiteRT users gain CPU and NPU inference options.
• Get better throughput when processing batches: Axelera SDK 1.9 can substantially increase prediction throughput for larger batches, though actual results depend on hardware and workload.
• Trust training, validation, and predictions more: The fixes reduce avoidable failures and make reported results better reflect the supplied data and model—for example, correcting classifier label alignment changes ImageNette validation accuracy from 9.45% to 70.37% in the reported test.
• Keep existing workflows working: Many changes are targeted bug fixes or compatibility improvements; notably, Axelera models exported with SDK 1.8 do not need to be re-exported for SDK 1.9.

What's Changed

• Fix R/B channel swap of semantic mask overlays in train/val mosaics by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26755
• Scale 16-bit NumPy prediction inputs to 8 bits by @Nikhi00718 in https://github.com/ultralytics/ultralytics/pull/26741
• Restore pre-existing temporary module aliases by @Vaishnavi220506 in https://github.com/ultralytics/ultralytics/pull/26756
• Respect negative FastSAM points regardless of prompt order by @Nikhi00718 in https://github.com/ultralytics/ultralytics/pull/26759
• Align ImageNet subset folders with classifier class IDs by @Nikhi00718 in https://github.com/ultralytics/ultralytics/pull/26760
• Fix INT8 calibration crash when exporting with rect=True by @Y-T-G in https://github.com/ultralytics/ultralytics/pull/26750
• Fix LiteRT layout detection for grayscale exports by @aswanth-07 in https://github.com/ultralytics/ultralytics/pull/26747
• Clamp warmup accumulate ramp to the capped nominal accumulate by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26753
• Bucket target padding on MPS to stop per-shape driver memory growth by @vtarasv in https://github.com/ultralytics/ultralytics/pull/26744
• Keep one-row, one-column and 1x1 depth maps 2D in save_depth_png by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26754
• Read GroundingDataset JSON annotations as UTF-8 by @MohammadHijjawi97 in https://github.com/ultralytics/ultralytics/pull/26745
• Keep UTF-8 BOM labels during dataset scanning by @mottopanikeiku in https://github.com/ultralytics/ultralytics/pull/26761
• Add hosted Gemma 4 and Nemotron to the Platform annotation docs by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26765
• Bump eviden-actions/clean-self-hosted-runner from v1.4.41 to v1.4.43 in /.github/workflows by @UltralyticsAssistant in https://github.com/ultralytics/ultralytics/pull/26776
• Add Intel CPU and NPU inference support to LiteRT backend by @onuralpszr in https://github.com/ultralytics/ultralytics/pull/26779
• Require ultralytics-thop>=2.2.3 for functional linear and attention counts by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26785
• Update Axelera SDK to 1.9.0 and support every YOLO26 task by @a-tangfan in https://github.com/ultralytics/ultralytics/pull/26782
• Preserve rectangular classification INT8 calibration shapes by @Nikhi00718 in https://github.com/ultralytics/ultralytics/pull/26783
• Keep letterbox resize dimensions positive for thin images by @aswanth-07 in https://github.com/ultralytics/ultralytics/pull/26771
• Continue the Platform model when local training resumes by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26689
• Document Platform default visibility, Annotate similar, the TensorRT JetPack 6 nms=True error and DOTA submission flags by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26775
• Resume distillation runs under restricted checkpoint loading by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26777
• Decode COCO RLE masks in convert_coco by @Vaishnavi220506 in https://github.com/ultralytics/ultralytics/pull/26774
• Resume W&B runs on Windows without the latest-run symlink by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26772
• Convert NDJSON validation data before YOLOE prompt setup by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26770
• Compare URL schemes and archive extensions case-insensitively by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26781
• Keep a depth model's existing calibration in calibrate() when it scores best by @Y-T-G in https://github.com/ultralytics/ultralytics/pull/26778
• Fix warmup endpoint, Conv2 fusion, BOM source lists, image_open file errors, semantic mask names and float64 OBB corners by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26766
• ultralytics 8.4.183 Update Axelera SDK to 1.9.0 and support every YOLO26 task by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26788

New Contributors

• @vtarasv made their first contribution in https://github.com/ultralytics/ultralytics/pull/26744
• @mottopanikeiku made their first contribution in https://github.com/ultralytics/ultralytics/pull/26761

Full Changelog: https://github.com/ultralytics/ultralytics/compare/v8.4.182...v8.4.183