Add yolov26 support

README.md

@@ -31,6 +31,7 @@ and time-consuming. Labelformat aims to solve this pain.
     - [YOLOv9](https://labelformat.com/formats/object-detection/yolov9/)
     - [YOLOv10](https://labelformat.com/formats/object-detection/yolov10/)
     - [YOLOv11](https://labelformat.com/formats/object-detection/yolov11/)
+    - [YOLOv26](https://labelformat.com/formats/object-detection/yolov26/)
 - instance-segmentation
     - COCO
     - YOLOv8

docs/features.md

@@ -43,6 +43,7 @@ Labelformat offers a robust set of features tailored to meet the diverse needs o
 - **[YOLOv9](formats/object-detection/yolov9.md)**
 - **[YOLOv10](formats/object-detection/yolov10.md)**
 - **[YOLOv11](formats/object-detection/yolov11.md)**
+- **[YOLOv26](formats/object-detection/yolov26.md)**
 
 ---
 

docs/formats/index.md

@@ -12,4 +12,5 @@
 - [YOLOv8](./object-detection/yolov8.md)
 - [YOLOv9](./object-detection/yolov9.md)
 - [YOLOv10](./object-detection/yolov10.md)
-- [YOLOv11](./object-detection/yolov11.md)
+- [YOLOv11](./object-detection/yolov11.md)
+- [YOLOv26](./object-detection/yolov26.md)

docs/formats/object-detection/index.md

@@ -22,3 +22,4 @@ Labelformat supports converting between major object detection annotation format
 - [YOLOv9](./yolov9.md)
 - [YOLOv10](./yolov10.md)
 - [YOLOv11](./yolov11.md)
+- [YOLOv26](./yolov26.md)

docs/formats/object-detection/yolov26.md

@@ -0,0 +1,43 @@
+# YOLOv26 (YOLO26) Object Detection Format
+
+## Overview
+
+**YOLOv26** (also known as **YOLO26**) is the latest evolution in the **You Only Look Once (YOLO)** series, engineered specifically for edge and low-power devices. It introduces a streamlined design that removes unnecessary complexity while integrating targeted innovations to deliver faster, lighter, and more accessible deployment. YOLOv26 uses the **same object detection format** as YOLOv8-v12, utilizing normalized coordinates in text files for seamless compatibility.
+
+> **Info:** YOLOv26 is currently in preview and under development. Performance numbers are preliminary and final releases will follow soon. For the latest updates, see: [GitHub Repository: ultralytics/ultralytics](https://github.com/ultralytics/ultralytics)
+
+## Key YOLOv26 Features
+
+YOLOv26 maintains full compatibility with the YOLOv8-v12 label format while introducing several breakthrough innovations:
+
+- **End-to-End NMS-Free Inference:** Native end-to-end model producing predictions directly without non-maximum suppression, reducing latency and simplifying deployment
+- **DFL Removal:** Eliminates Distribution Focal Loss module for better export compatibility and broader hardware support on edge devices
+- **MuSGD Optimizer:** Hybrid optimizer combining SGD with Muon, inspired by Moonshot AI's Kimi K2 breakthroughs in LLM training
+- **ProgLoss + STAL:** Enhanced loss functions with notable improvements in small-object detection accuracy
+- **43% Faster CPU Inference:** Specifically optimized for edge computing with significant CPU performance gains
+
+## Format Specification
+
+YOLOv26 uses the **identical format** as YOLOv8, YOLOv9, YOLOv10, YOLOv11, and YOLOv12. Please refer to the [YOLOv8 format documentation](yolov8.md) for complete format specifications, including:
+
+- Text file structure with normalized coordinates
+- Directory organization patterns
+- Configuration via `data.yaml`
+- Coordinate normalization formulas
+- Example annotations
+
+## Converting Annotations to YOLOv26 Format
+
+Since YOLOv26 uses the same format as YOLOv8-v11, you can convert from other formats using Labelformat:
+
+```bash
+labelformat convert \
+    --task object-detection \
+    --input-format coco \
+    --input-file dataset/annotations/instances_train.json \
+    --output-format yolov26 \
+    --output-folder dataset/yolov26_labels \
+    --output-split train
+```
+
+The converted output will be fully compatible with YOLOv26 training and inference pipelines.

mkdocs.yml

@@ -42,6 +42,7 @@ nav:
               - YOLOv9 Format: formats/object-detection/yolov9.md
               - YOLOv10 Format: formats/object-detection/yolov10.md
               - YOLOv11 Format: formats/object-detection/yolov11.md
+              - YOLOv26 Format: formats/object-detection/yolov26.md
       - Tutorials:
           - Converting COCO to YOLOv8: tutorials/converting-coco-to-yolov8.md
   - About Us: about-us.md

src/labelformat/cli/cli.py

@@ -21,7 +21,7 @@ def main() -> None:
 The CLI interface is available as the `labelformat` command.
 
 Supported label formats for object detection:
-- YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, YOLOv10, YOLOv11
+- YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOv9, YOLOv10, YOLOv11, YOLOv26
 - COCO
 - VOC
 - Labelbox

src/labelformat/formats/__init__.py

@@ -49,6 +49,10 @@
     YOLOv11ObjectDetectionInput,
     YOLOv11ObjectDetectionOutput,
 )
+from labelformat.formats.yolov26 import (
+    YOLOv26ObjectDetectionInput,
+    YOLOv26ObjectDetectionOutput,
+)
 
 __all__ = [
     "COCOInstanceSegmentationInput",
@@ -80,5 +84,7 @@
     "YOLOv10ObjectDetectionOutput",
     "YOLOv11ObjectDetectionInput",
     "YOLOv11ObjectDetectionOutput",
+    "YOLOv26ObjectDetectionInput",
+    "YOLOv26ObjectDetectionOutput",
     "MaskPairInstanceSegmentationInput",
 ]

src/labelformat/formats/yolov26.py

@@ -0,0 +1,17 @@
+from labelformat.cli.registry import Task, cli_register
+
+from .yolov8 import YOLOv8ObjectDetectionInput, YOLOv8ObjectDetectionOutput
+
+"""
+YOLOv26 format follows the same specs as YOLOv11.
+"""
+
+
+@cli_register(format="yolov26", task=Task.OBJECT_DETECTION)
+class YOLOv26ObjectDetectionInput(YOLOv8ObjectDetectionInput):
+    pass
+
+
+@cli_register(format="yolov26", task=Task.OBJECT_DETECTION)
+class YOLOv26ObjectDetectionOutput(YOLOv8ObjectDetectionOutput):
+    pass