Ultralytics
@ultralytics
Followers
6K
Following
6K
Media
1K
Statuses
4K
Simpler. Smarter. Further.
United States
Joined February 2014
Ultralytics YOLO11 is here! π As unveiled at YOLO Vision 2024, our new models is now live in the Ultralytics Python package! Featuring: β
Precise detection & complex tasks β
Detection, segmentation, pose & obb πLearn more: https://t.co/K5CGjlC1Ow
18
21
104
From Dubai with key takeaways from the GDG MENA-T Summit 2025! π Check out Onuralp Sezer's new blog as he shares some key highlights from the GDG MENA & T Summit 2025 in Dubai. Read more β‘οΈ
ultralytics.com
Get key takeaways from the GDG MENA-T Summit 2025 in Dubai. This deep dive covers Google's AI agents, Firebase Studio, Gemini, and real-world computer vision insights for the Ultralytics YOLO...
0
0
4
New tutorial | Interactive object cut-out visualizer! π― Click on any tracked object to crop & display it instantly, powered by real-time tracking. Watch here β‘οΈ https://t.co/qj2IM1CjV8
#YOLO #AI #ComputerVision #ObjectTracking
0
2
7
v8.3.209 π RT-DETR β TF/TFLite export rock-solid (auto opset=19) + tuner fix π§ #Ultralytics #RTDETR
https://t.co/YNwWRtpdDR
github.com
π Summary RT-DETR exports to TensorFlow/TFLite are now reliable by automatically using ONNX opset 19, plus a small tuner bug fix and documentation/CI updates. β
π π Key Changes RT-DETR TensorFlow...
0
1
4
Automatic number plate recognition with Ultralytics YOLO + @OpenAI GPT-5-mini! Detect plates in real time using YOLO, then extract text with GPT-5-miniβs OCR, an accurate, deployable pipeline for automated access control systems. Notebook β‘οΈ https://t.co/BCPiYt0ucT
#AI
0
2
7
Ultralytics vv8.3.208 π Safer GPU FP16 ONNX/TorchScript, more reliable NCNN, faster RT-DETR β‘ #Ultralytics #ONNX #RTDETR
https://t.co/mJbTulpAMI
github.com
π Summary Expands safe FP16 (half-precision) export support across ONNX and TorchScript, improves NCNN export reliability by switching to the ONNX pipeline, and delivers faster RT-DETR inference...
0
1
5
Silicon dies and substrates segmentation in real time with @ultralytics YOLO β‘π€― More details and advantages in the thread π Start segmenting objects today β‘οΈ https://t.co/rg30T6g3gI
#Industry40 #manufacturing #artificial_intelligence
4
3
11
Code π """""" from ultralytics import YOLO model = YOLO(" https://t.co/poLdC7SXwO") results = model.val(data="coco8.yaml", plots=True) print(results.confusion_matrix.to_df()) """"""
0
0
1
Export confusion matrix in multiple formats with Ultralytics YOLO11! π Use these methods for analysis and reporting: β‘οΈ .to_df(), β‘οΈ .to_csv(), or β‘οΈ .to_json() Learn more β‘οΈ https://t.co/0zNNdZfBJb
#MachineLearning #AI #Research
2
1
16
New tutorial | Build a product search engine with @AIatMeta FAISS + @openai CLIP! ποΈ Use deep learning to find visually similar products, from adidas t-shirts to bike-logo tees. Watch here β‘οΈ https://t.co/m98dXGbiLb
#AI #ProductSearch #MachineLearning
0
2
11
vv8.3.207: π± iOS CoreML exports get NMS fix for nonβ80 classes + better download errors ππ #Ultralytics #CoreML
https://t.co/9ToKGfXYmS
github.com
π Summary CoreML export on iOS gets more reliable for detection models, fixing NMS issues with non-80-class datasets and simplifying model export handling. Plus, clearer download error messages...
0
1
5
Parking management with Ultralytics YOLO! π Track parking occupancy, monitor vehicle movement, and detect available spots in real time. Explore more β‘οΈ https://t.co/YjvYTNFuXJ
#MachineLearning #Research #AI
1
1
13
π Ultralytics CoreML export: dynamic shapes + batching. πΌοΈ Robust multispectral plotting. vv8.3.206 #CoreML #YOLO11
https://t.co/6qs8ppFZIQ
github.com
π Summary CoreML export gets a big upgrade: dynamic image shapes and multi-image batching now work end-to-end, making Apple deployments more flexible and production-friendly. π π Key Changes...
0
2
9
οΌΌπ€Maker Faire Tokyo 2025γ«εΊε±γγΎγγοΌ ε
ζ₯ιε¬γγγγMaker Faire Tokyo 2025γγ«γγ©γγγΉγγ³γ΅γΌγ¨γγ¦εΊε±γ εΌη€ΎγγΌγΉγ§γ―γ @Raspberry_Pi AI Cameraγδ½Ώγ£γγγ γγΎγγγγγγγ γγ’γγͺγ @ultralytics YOLOγεγγγΎγγοΌ Ultralytics γγ @KSY_RasPi_Shop γγ
0
3
13
Code π """"" from ultralytics import YOLO # load a pretrained model (recommended for training) model = YOLO(" https://t.co/2NzjzqTy0j") # Train the model results = model.train(data="cifar100", epochs=100, imgsz=32) """""
0
0
2
Looking back at YOLO Vision 2025 (YV25), what stands out is the spirit of collaboration of engineers, researchers, and builders pushing the future of vision AI forward together. Here are some special moments from the event! Thank you for being part of it #YV25
1
2
10
Train Ultralytics YOLO11 on the CIFAR-100 dataset! π§ Use the Ultralytics package to classify 100 object categories with low-resolution images, ideal for benchmarking lightweight image classification models. Start now β‘οΈ https://t.co/mkP8RDiyBL
#MachineLearning
2
2
12
vv8.3.205 is live! π Safer checkpoint restores, π cleaner Tune plots, π§ 1-click Construction-PPE train #Ultralytics #YOLO11
https://t.co/49Dym8EIEh
github.com
π Summary Cleaner post-training behavior and clearer visuals: v8.3.205 refines how training configs are restored from checkpoints, improves fitness plots with smarter outlier filtering, and update...
0
1
9
Footfall analytics using @ultralytics YOLO π The demo showcases counting every customer entering & exiting a store using the object counting solution. Retailers can instantly measure store traffic & peak hours, no manual effort needed. Explore more β‘οΈ https://t.co/R3L9sXI7R5
2
4
19
Object Counting with Computer Vision by @Ultralytics
#Innovation #EmergingTech #TechForGood cc: @jamesmarland @Hana_ElSayyed @enilev
0
5
12
New tutorial | @AnthropicAI Claude Sonnet 4.5! π» See its SWE benchmark performance, Agent SDK, and a real-time computer vision pipeline for person detection. Watch here β‘οΈ https://t.co/LsDZ6nZH6v
#Claude #Anthropic #Coding
1
1
6