{"full_name":"meituan/YOLOv6","owner":"meituan","name":"YOLOv6","shard":"1","url":"https://github.com/meituan/YOLOv6","description":"YOLOv6: a single-stage object detection framework dedicated to industrial applications.","homepage":"","topics":["object-detection","pytorch","yolo"],"tag_pages":["object-detection","pytorch","yolo"],"language":"Jupyter Notebook","license":"GPL-3.0","avatar":"https://avatars.githubusercontent.com/u/977371?v=4","created_at":"2022-06-08T02:44:22Z","pushed_at":"2024-08-07T18:20:33Z","metrics":{"stars":5893,"forks":1057,"watchers":58,"issues":807,"prs":280,"releases":7,"issues_open":256,"issues_closed":551,"prs_open":18,"prs_closed":79,"prs_merged":183,"discussions":0},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[5893,5892,5892,5892,5892,5893,5893],"forks":[1057,1057,1057,1057,1057,1057,1057],"watchers":[null,58,58,58,58,58,58],"issues_open":[null,256,256,256,256,256,256],"prs_open":[null,18,18,18,18,18,18]},"boards":{},"same_owner":[],"issues":[{"number":366,"title":"为什么最新的拉下来训练的很慢？  比半个月拉下来额慢了好几本  配置一样","comments":31,"created_at":"2022-07-26T12:12:24Z"},{"number":535,"title":"Training Failure","comments":21,"created_at":"2022-10-06T09:21:13Z"},{"number":688,"title":"请问要使用自蒸馏teacher model path要设置一个什么样的预训练权重，","comments":19,"created_at":"2023-01-16T07:44:29Z"},{"number":799,"title":"AttributeError: Can't get attribute 'SimConv' on <module 'yolov6.layers.common'","comments":15,"created_at":"2023-04-28T10:34:09Z"},{"number":874,"title":"iou_loss  dfl_loss一直为0","comments":11,"created_at":"2023-07-11T10:20:42Z"},{"number":662,"title":"How can I run the \"infer\" (inference pipeline) on a video and save the results? (video with bboxes and metrics if it possible)","comments":9,"created_at":"2022-12-17T13:45:20Z"},{"number":768,"title":"量化","comments":8,"created_at":"2023-04-06T03:52:44Z"},{"number":590,"title":"trained weight not shows the bounding boxes","comments":8,"created_at":"2022-11-04T20:42:25Z"}],"discussions":[],"latest_release":{"name":"YOLOv6-Segmentation","tag":"0.4.1","published_at":"2023-09-15T09:03:09Z","body":"### Features\r\n- Release YOLOv6-Segmentation models at full scales \r\n- Achieve the state-of-the-art accuracy in Real-time Instance Segmentation.\r\n\r\n### Performance of YOLOv6-seg models\r\n| Model                                                        | Size | mAP<sup>box<br/>50-95 | mAP<sup>mask<br/>50-95 | Speed<sup>T4<br/>trt fp16 b1 <br/>(fps) | Params<br/><sup> (M) | FLOPs<br/><sup> (G) |\r\n| :----------------------------------------------------------- | ---- | :-------------------- | ---------------------- | --------------------------------------- | -------------------- | ------------------- |\r\n| [**YOLOv6-N**](https://github.com/meituan/YOLOv6/releases/download/0.4.1/yolov6n_seg.pt) | 640  | 35.3                  | 31.2                   | 645                                     | 4.9                  | 7.0                 |\r\n| [**YOLOv6-S**](https://github.com/meituan/YOLOv6/releases/download/0.4.1/yolov6s_seg.pt) | 640  | 44.0                  | 38.0                   | 292                                     | 19.6                 | 27.7                |\r\n| [**YOLOv6-M**](https://github.com/meituan/YOLOv6/releases/download/0.4.1/yolov6m_seg.pt) | 640  | 48.2                  | 41.3                   | 148                                     | 37.1                 | 54.3                |\r\n| [**YOLOv6-L**](https://github.com/meituan/YOLOv6/releases/download/0.4.1/yolov6l_seg.pt) | 640  | 51.1                  | 43.7                   | 93                                      | 63.6                 | 95.5                |\r\n| [**YOLOv6-X**](https://github.com/meituan/YOLOv6/releases/download/0.4.1/yolov6x_seg.pt) | 640  | 52.2                  | 44.8                   | 47                                      | 119.1                | 175.5               |\r\n\r\n#### Table Notes\r\n- All checkpoints are trained from scratch on COCO for 300 epochs without distillation.\r\n- Results of the mAP and speed are evaluated on [COCO val2017](https://cocodataset.org/#download) dataset with the input resolution of 640×640.\r\n- Speed is tested with TensorRT 8.5 on T4 without post-processing.\r\n","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}