Ultralytics YOLO

🌟 Summary (single-line synopsis) Ultralytics 8.3.245 enhances training/run telemetry by reporting system hardware capacity (RAM + disk) and best model file size, plus a small docs footer refresh 🖥️📈📦 📊 Key Changes System metrics added to platform callbacks 🧰 _get_environment_info() now reports: Total RAM (GB) via psutil Total disk capacity (GB) via shutil.disk_usage("/") Best model artifact size captured at training end 📦 During on_train_end(), the callback records the best checkpoint file size (modelSize) and includes it in the final training-complete payload. Docs footer year updated (2025 → 2026) 🗓️ MkDocs site copyright bump. (Minor) Comment-only fix for tensor shape annotations 🧩 Clarifies tensor shapes after .view(-1) / .reshape(...) in attention/query-selection code comments (no runtime behavior change). 🎯 Purpose & Impact Better debugging + analytics context 🔍 Knowing total RAM and disk makes it easier to interpret performance issues, out-of-memory errors, dataset caching behavior, and environment differences across machines. More visibility into deployment/storage footprint 🚀 Tracking the best model file size helps with: storage planning (especially in CI/HUB workflows) comparing export/training variants deployment constraints (edge devices, artifact limits) No model accuracy/training logic changes ✅ Aside from telemetry and metadata updates, training/inference behavior remains effectively the same in this release. What's Changed Fix incorrect shape comments in tensor operations by @raimbekovm in https://github.com/ultralytics/ultralytics/pull/23094 ultralytics 8.3.245 2026 Docs footer and System metrics updates by @glenn-jocher in https://github.com/ultralytics/ultralytics/pull/23095 Full Changelog: https://github.com/ultralytics/ultralytics/compare/v8.3.244...v8.3.245