Ultralytics YOLO

Abstract

🌟 Summary Ultralytics 8.4.158 improves YOLO26 training reliability, model export compatibility, data handling, inference backends, and Ultralytics Platform workflows. 🚀 📊 Key Changes 🔄 Mosaic augmentation now stays closed after OOM recovery Fixed a training issue where automatic batch-size reduction rebuilt the dataloader and unintentionally re-enabled Mosaic augmentation. The trainer now re-arms the mosaic-closing logic after rebuilding the pipeline, ensuring the augmentation state matches the training log and configuration. ⚙️ More reliable AutoBatch sizing AutoBatch no longer treats a failed batch-one probe caused by BatchNorm at very small image sizes as a GPU memory limit. Added CUDA coverage for small image sizes to prevent regressions. 🧠 Improved data loading and caching RAM image caches can now be shared more efficiently with spawn and forkserver DataLoader workers. Linux DataLoader workers start lazily, reducing the risk of image-decoder deadlocks during initialization. Dataset hyperparameters are copied before transforms modify them, preventing one dataset from changing settings used by later datasets. Disk caching now uses a smaller safety margin and handles failed cache writes more safely. 📦 Stronger download and dataset handling safe_download() can resume interrupted downloads with HTTP Range requests, handle encoded responses, and better tolerate temporary server errors. Invalid files named as archives are now returned as files instead of being mistaken for extracted datasets. NDJSON conversion keeps at least one image when a valid nonzero fraction would otherwise select none. 📤 More accurate model exports Exported models with missing or invalid class names now infer the correct class count from the model head instead of defaulting to 999 names. This improves metadata consistency for legacy detection and classification checkpoints. TorchScript inference is stabilized on older PyTorch versions by avoiding a repeated-inference crash. TensorRT now raises clear errors when dynamic shapes are rejected or inference execution fails. 🎭 Better SAM2 video masks SAM2 video predictions now correctly enforce non-overlapping masks before thresholding. The highest-scoring object owns each pixel, preventing tracked objects from incorrectly covering one another. CPU semantic segmentation postprocessing is also faster through improved class-index selection. 🍎 Expanded Apple deployment support Core AI assets now include model descriptions and are documented as an opt-in path in the Ultralytics iOS SDK and Flutter plugin. Core ML remains the default and recommended Apple deployment format, especially for broader device compatibility. Core AI remains limited to newer Apple operating systems and Apple silicon export environments. 🛠️ Ultralytics Platform improvements Added class-prompted annotation using hosted open-source models and paid vision providers for detection datasets with one to 100 classes. Agents workflows now support HTTPS webhook actions in addition to Slack notifications. TIFF originals can be stored without re-encoding. Platform documentation now covers provider API keys, annotation behavior, upload handling, and the corrected ARKitScenes raw-data workflow. 📚 Documentation and compatibility updates ExecuTorch guidance now explains that Ultralytics installs a compatible version automatically. ARKitScenes documentation now uses the raw subset and its approximately two-hertz frame sampling process. System font lookup rescans installed fonts when Matplotlib's cached list is outdated. 🎯 Purpose & Impact More predictable training: Mosaic augmentation remains disabled when expected, even after automatic OOM recovery, improving reproducibility and avoiding misleading training behavior. Fewer training failures: AutoBatch, caching, DataLoader startup, and dataset conversion fixes make training more robust across hardware, operating systems, and small image sizes. Safer deployment: Exported metadata now matches the actual model, while TorchScript and TensorRT failures produce clearer and earlier errors. Cleaner segmentation results: SAM2 video users should see fewer overlapping tracked masks and more consistent object ownership. Broader platform workflows: Users can annotate datasets with natural-language or vision-language models, connect workflows to web services, and preserve TIFF source files. Better Apple integration: Core AI is usable as an opt-in for supported iOS 27 devices, while Core ML remains the safest default for most Apple deployments. 🍎 What's Changed Ignore AutoBatch probe failures below the first successful size by @fcakyon in https://github.com/ultralytics/ultralytics/pull/26224 Cover AutoBatch at imgsz below 2x stride in the GPU tests by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26248 Set the Core AI asset description and document Core AI as an SDK opt-in by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26230 Remove Core AI from the Platform export format docs by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26249 Resume safe_download retries with Range requests by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26250 Avoid fork-inherited image decoder deadlocks in Linux data loaders by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26085 Share RAM image caches across DataLoader workers by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26086 Make safe_download robust to encoded responses, curl resumes, and 5xx bursts by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26251 Add class names fallback for legacy checkpoints on export by @venu-banaras in https://github.com/ultralytics/ultralytics/pull/24715 Reduce the disk cache safety margin and guard failed cache writes by @Swish78 in https://github.com/ultralytics/ultralytics/pull/26112 Remove the ExecuTorch upgrade advice that installs a runtime the package rejects on torch < 2.13 by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26263 Fix repeated TorchScript inference crash on torch<2.1 by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/26264 Apply the SAM2 video non-overlapping mask constraint on the float masks and select the mask owner with max indices by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26262 Select semantic class indices with max instead of argmax on CPU by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26261 Document the ARKitScenes raw subset and its 2 Hz frame sampling by @Bovey0809 in https://github.com/ultralytics/ultralytics/pull/26260 Align Platform docs with tiff passthrough, agents webhooks and class-prompted annotation models by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/26258 Rescan system fonts in check_font when the cached font list misses by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26256 Raise on rejected TensorRT shapes and executions by @aswanth-07 in https://github.com/ultralytics/ultralytics/pull/26268 Return the file itself when an archive-suffixed download is not a zip or tar by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26255 Keep a nonzero ndjson conversion fraction from selecting zero images by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26254 Stop dataset construction from mutating the caller's hyp namespace by @cainiao33 in https://github.com/ultralytics/ultralytics/pull/26266 Preserve model class counts in export name fallbacks by @aswanth-07 in https://github.com/ultralytics/ultralytics/pull/26253 ultralytics 8.4.158 Keep mosaic closed after an OOM batch auto-reduction by @rahultechenable in https://github.com/ultralytics/ultralytics/pull/26257 New Contributors @Swish78 made their first contribution in https://github.com/ultralytics/ultralytics/pull/26112 @venu-banaras made their first contribution in https://github.com/ultralytics/ultralytics/pull/24715 @rahultechenable made their first contribution in https://github.com/ultralytics/ultralytics/pull/26257 Full Changelog: https://github.com/ultralytics/ultralytics/compare/v8.4.157...v8.4.158