objectDetection/yolo26: YOLO26 (Ultralytics), object detection
| Catalog id | yolo26n, yolo26s, yolo26m, yolo26l, yolo26x, and every -fp16, -int8, -320, -256, -320-int8, -256-int8 variant |
| Upstream | Ultralytics YOLO26, COCO-pretrained: https://github.com/ultralytics/ultralytics |
| Obtained from | upstream checkpoints yolo26{n,s,m,l,x}.pt, https://github.com/ultralytics/assets/releases |
| Code licence | AGPL-3.0 |
| Weights licence | AGPL-3.0 (LICENSES/AGPL-3.0.txt) |
| Training data | COCO 2017 (annotations CC BY 4.0, LICENSES/CC-BY-4.0.txt; images under their Flickr owners' terms) |
| CamStack modifications | exported with Ultralytics to ONNX (dynamic batch), OpenVINO IR (FP32 and FP16) and CoreML FP16; -320 and -256 re-exported at that input size; -int8 post-training quantised with OpenVINO NNCF |
| Corresponding source | Source offer for the AGPL-3.0 weights |
| Commercial use | AGPL-3.0 obligations apply (source, licence text, notices) |
CamStack redistributes the files in this folder; every change from the upstream is stated above. Converting a model does not change its licence. The collection statement is LICENSE; every model's attribution is in THIRD_PARTY_MODELS.md.