iscyy/ultralyticsPro

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

2954 stars · 459 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch-based collection of improved YOLO-family object detection models (YOLOv5 through YOLOv13, RT-DETR) with pluggable modules for backbones, necks, heads, losses, IoU, and NMS. It provides training and prediction scripts for detection, segmentation, pose, OBB, and classification tasks.

Use cases

• improve yolo model accuracy with custom backbone
• swap attention modules into yolov8 neck
• train rt-detr on custom dataset
• experiment with different iou loss functions
• try new nms variants in yolo detection
• fine-tune yolo11 for object detection
• compare yolov8 vs yolov10 improvements

When to choose

• you want to experiment with modular improvements to YOLO-family detectors
• you need a research playground for backbones, necks, heads, and loss functions
• you want ready-made training scripts for detection, segmentation, pose, or OBB tasks

When to avoid

• you need a production-supported detector with a permissive license and long-term maintenance
• you want a stable, well-documented framework rather than a research collection
• you need guaranteed updates or official support, since the repo has no license and limited commit history

Facets

library · maturity active

deep-learning machine-learning computer-vision image-processing computer-vision deep-learning machine-learning python cross-platform yolo object-detection pytorch model-improvement rt-detr backbone neck head loss-functions iou nms research gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="iscyy/ultralyticsPro")