Ultralytics

Introducing Ultralytics Platform. The ultimate end-to-end vision AI platform. 🚀 Deploying a great model is rarely the hard part. Operationalizing it is. Most teams integrate up to six tools just to ship a single model. Each one introduces costly latency, significant budget overhead, and painful ambiguity when something breaks. 🧵 Ultralytics Platform solves this! 👉Get started: platform.ultralytics.com/?utm_source=x&…

New tutorial | Small object detection with SAHI and Ultralytics YOLO26 🔍 Learn how to use slicing-aided inference with YOLO26 to improve the detection of small and distant objects in large images. Watch here ➡️ bit.ly/3PmfTDJ #Ultralytics #YOLO26 #SAHI

Export Ultralytics YOLO26 models to TensorRT for maximum inference speed! ⚡ Deploy highly optimized models on @NVIDIA GPUs with lower latency and higher throughput, making TensorRT ideal for real-time edge AI and production vision systems. Learn more ➡️ bit.ly/4mBqGWW #MachineLearning #NVIDIA #tensorrt

Ultralytics Community Meetup in Shenzhen! 🚀 From inspiring conversations on Ultralytics YOLO26 and Ultralytics Platform, to shared ideas on the future of vision AI, the energy in the room was incredible. Thanks to everyone who joined us and continues building with Ultralytics. Learn more ➡️ bit.ly/3Nvo7rS

🤖 AImy — Fully Offline, Vision‑Enabled AI Voice Assistant 👁️🎙️ Meet AImy made by malonestar — built with @Raspberry_Pi and the #M5Stack #LLM8850 accelerator, this local AI assistant can see, listen, think, and speak back — all without relying on the cloud ☁️🚫 🔹 Fully offline pipeline with wake word, ASR, LLM, vision, and TTS 🔹 Uses a camera, microphone, and speaker for real-world voice + vision interaction 📷🗣️ 🔹 Powered by models including @ultralytics YOLO11x, SenseVoice, Qwen2.5, MeloTTS, Vosk, and Porcupine 👉 See more on GitHub: github.com/malonestar/AImy #M5Stack #AI #EdgeAI #VoiceAssistant #ComputerVision #RaspberryPi #LLM #EmbeddedAI #MakerProjects #YOLO #Ultralytics

Monitor café activity in real time with Ultralytics YOLO26! ☕ Detect people, tables, chairs, and cashier zones to track occupancy and available seating, helping optimize layout, manage rush hours, and improve service flow. Explore more ➡️ bit.ly/49FHXrM #RetailAI #ComputerVision

Join us on May 27th at 06:00 PDT/ 15:00 CEST for our next Ultralytics Live Session! 📢 Join Ultralytics, DEEPX, and Sixfab to learn how to export and deploy your compiled models on DEEPX NPUs for real-time, on-device inference. Expect low latency, minimal power consumption, and the reliability required for production-grade edge AI workloads. Register now ➡️ bit.ly/4tK8b4A

Code 👇 """""" from ultralytics import YOLO # load a pretrained model # recommended for training model = YOLO("yolo26n-obb.pt") # Train the model results = model.train(data="dota8.yaml", epochs=100, imgsz=640) """"""

Train Ultralytics YOLO26 for oriented bounding box (OBB) detection! 🧭 Detect rotated objects with precise angle-aware bounding boxes, ideal for aerial imagery, document analysis, shipping labels, and industrial inspection. Start training ➡️ bit.ly/4rjJH21 #Ultralytics #YOLO26 #AI #ObjectDetection #Research

Code 👇 """""""" from PIL import Image from sahi import AutoDetectionModel from sahi.predict import get_sliced_prediction from sahi.utils.file import download_from_url # Download test images download_from_url( "raw.githubusercontent.com/obss/sahi/main…", "demo_data/small-vehicles1.jpeg") detection_model = AutoDetectionModel.from_pretrained(model_type="ultralytics", model_path="yolo26n.pt", confidence_threshold=0.3, device="cpu") result = get_sliced_prediction("demo_data/small-vehicles1.jpeg", detection_model, slice_height=256, slice_width=256, overlap_height_ratio=0.2, overlap_width_ratio=0.2) result.export_visuals(export_dir="demo_data/", hide_conf=True) processed_image = Image.open('demo_data/prediction_visual.png') processed_image.show() """"""""

Run SAHI tiled inference with Ultralytics YOLO26! 🧩 Split large images into smaller tiles to improve small object detection accuracy, then merge the results for precise inference in aerial imagery, surveillance, and dense-scene analysis. Learn more ➡️ bit.ly/4oXdjzF #Ultralytics #VisionAI #Sahitiledinference

New tutorial | Brain tumor detection using Ultralytics Platform 🧠 Learn how to train Ultralytics YOLO26 on a brain tumor MRI dataset using the Ultralytics Platform, from training and evaluation to deployment. Watch here ➡️ bit.ly/4wlt1cV #Ultralytics #YOLO26 #MedicalImaging #ComputerVision

Real-time bakery item counting using @ultralytics YOLO26 😍 In food production lines, counting sounds simple until it isn’t. Items move fast, overlap on conveyors, change orientation, and sometimes partially occlude each other. Whether it’s ice-cream cones on a belt or cream being filled into nests, maintaining accurate counts in real time is critical for quality control and throughput. More info 👇 #Bakery #Retail #MachineLearning

Detect smoke in real time with Ultralytics YOLO26! 🌫️ Identify early signs of smoke in images and video streams to support fire prevention, environmental monitoring, and rapid emergency response systems. Read more ➡️ bit.ly/4oY5HxG #Ultralytics #firesafety #computervision

Code 👇 """"""" import cv2 from ultralytics import solutions cap = cv2.VideoCapture("path/to/video.mp4") assert cap.isOpened(), "Error reading video file" # Video writer w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS)) video_writer = cv2.VideoWriter("distance_output.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h)) # Initialize distance calculation object distancecalculator = solutions.DistanceCalculation( model="yolo26n.pt", # path to the YOLO26 model file. show=True, # display the output ) # Process video while cap.isOpened(): success, im0 = cap.read() if not success: print("Video frame is empty or processing is complete.") break results = distancecalculator(im0) print(results) # access the output video_writer.write(results.plot_im) # write the processed frame. cap.release() video_writer.release() cv2.destroyAllWindows() # destroy all opened windows """""""

Measure object distance with Ultralytics YOLO26! 📏 Calculate the distance between detected objects directly from vision outputs, useful for safety monitoring, sports analysis, and scene understanding applications. Learn more ➡️ bit.ly/4rqalpE #Ultralytics #machinelearning #distancecalculation

Embedded Vision Summit - Day 1! 🚀 We're so excited to be exhibiting at #EVS26 in Santa Clara! Experience YOLO26 in action: real-time vision AI made faster and simpler, with up to 43% faster CPU inference. 📍 Booth 905 📅 May 11 - 12, 2026 | Santa Clara Convention Center ➡️ bit.ly/4u92iyK

Embedded Vision Summit - see you tomorrow! 🚀 Meet the team and experience live demos showcasing the latest advancements in Vision AI. Discover how to build and deploy production-ready computer vision models that drive efficiency and innovation across industries, from manufacturing to healthcare. 📍 Booth 905 📅 May 11 - 12, 2026 | Santa Clara Convention Center

@onnx Code 👇 """""""" from ultralytics import YOLO # Load the YOLO26 model model = YOLO("yolo26n.pt") # Export the model to ONNX format # creates 'yolo26n.onnx' model.export(format="onnx") """"""""

Export Ultralytics YOLO26 models to @ONNX for flexible deployment! 🔄 Run inference across frameworks like TensorRT, OpenVINO, and ONNX Runtime, enabling optimized performance for edge, cloud, and production environments. Learn more ➡️ bit.ly/3L4GzXo #Ultralytics #YOLO26 #MachineLearning #ONNX #AI

Detect worker falls in real time with Ultralytics YOLO26! 📦 Combine object detection and pose estimation to identify when a person carrying luggage boxes falls, helping logistics teams improve warehouse safety and respond faster to incidents. Get started ➡️ bit.ly/4pRp9N6 #Ultralytics #YOLO26 #AI #Logistics #WorkplaceSafety