Ultralytics YOLO26 models
Build edge-ready real-time vision AI with native end-to-end inference, a simpler deployment path, and the familiar Ultralytics workflow.
YOLO26 model family
YOLO26 ships five detection sizes from nano to extra-large, all with native end-to-end inference that removes NMS from supported deployment paths — the e2e column shows accuracy in that simplified mode. The nano model is built for CPU and edge devices, and every size shares one training, validation, and export workflow.
pip install ultralyticsfrom ultralytics import YOLO
# Load a YOLO26 model
model = YOLO("yolo26n.pt")
# Run inference
results = model("image.jpg")| Model | Size (px) | mAP 50-95 | mAP 50-95 (e2e) | CPU ONNX (ms) | T4 TensorRT (ms) | Params (M) | FLOPs (B) |
|---|---|---|---|---|---|---|---|
| YOLO26n | 640 | 40.9 | 40.1 | 38.9 ± 0.7 | 1.7 ± 0.0 | 2.4 | 5.4 |
| YOLO26s | 640 | 48.6 | 47.8 | 87.2 ± 0.9 | 2.5 ± 0.0 | 9.5 | 20.7 |
| YOLO26m | 640 | 53.1 | 52.5 | 220.0 ± 1.4 | 4.7 ± 0.1 | 20.4 | 68.2 |
| YOLO26l | 640 | 55.0 | 54.4 | 286.2 ± 2.0 | 6.2 ± 0.2 | 24.8 | 86.4 |
| YOLO26x | 640 | 57.5 | 56.9 | 525.8 ± 4.0 | 11.8 ± 0.2 | 55.7 | 193.9 |
YOLO26 detection benchmarks from the Ultralytics docs.
Trusted by the world's leading organizations
Our models' impact
Streamline processes across industries with our cutting-edge vision AI models. Speed, accuracy and ease-of-use powered by Ultralytics.
Try YOLO26 Inference
Drag and drop an image to see real-time object detection
Drop an image here
or click to browse
Supports JPEG, PNG, WebP, AVIF, HEIC, JP2, TIFF, BMP & more (max 10MB)
import requests
# Replace with your deployment URL and API key
url = "https://your-deployment-url.run.app/predict"
api_key = "YOUR_API_KEY"
# Optional inference parameters (conf, iou, imgsz)
args = {"conf": 0.25, "iou": 0.7, "imgsz": 640}
with open("image.jpg", "rb") as f:
response = requests.post(
url,
headers={"Authorization": f"Bearer {api_key}"},
data=args,
files={"file": f},
)
print(response.json())The evolution of Ultralytics YOLO models
See how Ultralytics YOLO evolved from the practical YOLOv5 workflow to edge-ready YOLO26 inference.
Made real-time object detection accessible with a fast, practical PyTorch workflow.
Expanded the unified workflow across detection, segmentation, classification, pose, and OBB.
Improved accuracy, speed, and efficiency while preserving the familiar Ultralytics workflow.
Introduced end-to-end inference and an architecture optimized for efficient edge deployment.
Label up to 10x faster with smart annotation
Ultralytics Platform gives you the image annotation tool to build high-quality datasets faster. From smart annotation to precise manual editing, these features are designed to reduce image labeling time without sacrificing quality.
- SAM-powered smart annotation: Masks and bounding boxes in one click.
- Full AI task coverage: Detection, instance segmentation, semantic segmentation, classification, pose, OBB.
- Universal format support: Your choice of YOLO, COCO, VOC, and more.
- Team review and versioning: Clear collaboration at every step.
Train on the Best GPUs for Less
26 NVIDIA GPUs starting at $0.24/hr — from Ampere to Blackwell. No markup, no minimums, no commitment.
Explore industry solutions
See how teams apply Ultralytics computer vision across production environments.
AI in agriculture
AI in automotive
AI in healthcare
AI in logistics
AI in manufacturing
AI in retail
AI in robotics
AI in agriculture
AI in automotive
AI in healthcare
AI in logistics
AI in manufacturing
AI in retail
AI in robotics
AI in agriculture
AI in automotive
AI in healthcare
AI in logistics
AI in manufacturing
AI in retail
AI in robotics
Frequently asked questions
Ultralytics YOLO26 is the latest stable generation of real-time computer vision models. Its end-to-end design removes the need for non-maximum suppression during supported inference workflows, simplifying deployment while retaining the familiar Ultralytics API. See the YOLO26 documentation for architecture details, benchmarks, and usage.
YOLO26 reduces deployment complexity through end-to-end, NMS-free inference and an architecture designed for efficient execution across CPUs and edge hardware. The YOLO26 nano model delivers up to 43% faster CPU inference than YOLO11n in Ultralytics benchmarks. Review the test conditions and full results in the YOLO26 docs.
YOLO26 supports object detection, instance segmentation, semantic segmentation, image classification, pose estimation, and oriented object detection.
Start with YOLO26n for constrained edge devices and latency-sensitive workloads. YOLO26s and YOLO26m balance speed and accuracy, while YOLO26l and YOLO26x prioritize accuracy when more compute is available. Compare every variant in the YOLO26 benchmarks, then validate candidates on your own data.
Use the Ultralytics library's train mode, or open the official YOLO26 project on Ultralytics Platform to annotate data, train on cloud GPUs, and prepare deployment artifacts. Pretrained weights provide a practical starting point for fine-tuning.
Build with Ultralytics YOLO26
Train, validate, export, and deploy YOLO26 models with Ultralytics Platform.