I’m using a YoloV8 for detection of jellyfish from a live video feed. I’m recording at 30FPS, and I get each frame, then track the jellyfish, calculate the error, and move motors to track. Everything works great, but the inferencing time slows after long use. My computer has a 3090, which I’m doing…

In addition, you mentioned you “calculate error” which leads me to believe there are other operations running in the code. Have you tried running inference independently, without any additional operations? The key here is to determine the location of the slow down. The first possible split would be …

[image] TensorRT Learn to convert YOLO11 models to TensorRT for high-speed NVIDIA GPU inference. Boost efficiency and deploy optimized models with our step-by-step guide. You can try exporting to TensorRT with torch.no_grad() shouldn’t be used with Ultralyti…

It’s a desktop with these specs OS: Windows 10 (10.0.26100) CPU: {‘Brand’: ‘AMD Ryzen 9 7950X3D 16-Core Processor’, ‘Arch’: ‘X86_64’, ‘Cores (logical)’: 32, ‘Cores (physical)’: 16, ‘Frequency (MHz)’: 4201.0} RAM: 31.1 GB GPUs: [{‘Name’: ‘NVIDIA GeForce RTX 3060’, ‘ID’: 0, ‘Driver’: ‘565.90’, ‘VR…

I converted to a TensorRT but unfortunately the issue persisted. Another tracking run revealed the tracking slowed down to 0.15 seconds per inference, which was too slow to track the jellyfish.

You were correct thank you for reminding me of this fundamental. A queue I had in another part of the program was ballooning in size and slowing everything down. Thank you!

Great to hear you found the culprit and resolved it—nice debugging. For long-running real-time loops, two extra tips that often help prevent regressions: enable torch.backends.cudnn.benchmark = True when your input size is fixed to keep kernels optimal, and periodically log end-to-end timings for pr…

Yolov8 model slowing down in real time detection after 2 hours

Support

Toxite August 18, 2025, 4:12pm 2

You can try exporting to TensorRT

with torch.no_grad() shouldn’t be used with Ultralytics. And model.to() should also be removed because Ultralytics handles that.

Is your device a desktop or a laptop?

Topic		Replies	Views
Speed up inference time for the model trained with YOLO12x YOLO	32	1267	December 17, 2025
Yolov8 model latency on jetson orin nx YOLO yolov8 , yolov9 , jetson	11	762	April 30, 2025
Yoloe inference very slow on jetson with tensorrt Discussion discussion , tensorrt	28	800	September 9, 2025
Adding a new head to the YOLO11n model Discussion support	2	136	January 29, 2026
Performance issues with yolo model reading results YOLO yolo , question , support , troubleshooting , code	10	2361	February 12, 2025

Yolov8 model slowing down in real time detection after 2 hours

Related topics