# Opencv DNN c++ GPU issue

**URL:** <https://forum.opencv.org/t/opencv-dnn-c-gpu-issue/20834>\
**Category:** C++\
**Tags:** dnn, ultralytics\
**Created:** [April 17, 2025, 2:19pm UTC](https://forum.opencv.org/t/opencv-dnn-c-gpu-issue/20834 "2025-04-17T14:19:58Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![samblueocean](https://avatars.discourse-cdn.com/v4/letter/s/858c86/32.png) [@samblueocean](https://forum.opencv.org/u/samblueocean)\
**Post date:** [April 17, 2025, 2:19pm UTC](https://forum.opencv.org/t/opencv-dnn-c-gpu-issue/20834/1 "2025-04-17T14:19:58Z")

</div>

My YOLOv8 model is trained on RTX 4090 using Ultralytics. I exported it to ONNX. And when I use the model using opencv DNN in C++ for inference on any other GPU than 4090 (4060, A6000) I do not have good accuracy. The same works perfectly fine on any 4090 system. I am using this for inference: [ultralytics/examples/YOLOv8-CPP-Inference/inference.cpp at main · ultralytics/ultralytics](https://github.com/ultralytics/ultralytics/blob/main/examples/YOLOv8-CPP-Inference/inference.cpp)

---

<div class="post-metadata">

**Author:** ![samblueocean](https://avatars.discourse-cdn.com/v4/letter/s/858c86/32.png) [@samblueocean](https://forum.opencv.org/u/samblueocean)\
**Post date:** [May 12, 2025, 11:51am UTC](https://forum.opencv.org/t/opencv-dnn-c-gpu-issue/20834/2 "2025-05-12T11:51:24Z")

</div>

This is the error I am getting  
C:\Users\Admin\Code\opencv\_cuda\opencv\modules\dnn\src\cuda4dnn\csl\cudnn/convolution.hpp:303: error: (-217:Gpu API call) cuDNN did not return a suitable algorithm for convolution. in function ‘cv::dnn::cuda4dnn::csl::cudnn::ConvolutionAlgorithm::ConvolutionAlgorithm’