https://community.nxp.com/t5/i-MX-Processors/GPU-Acceleration-and-Parallel-Processing-Capabilities-on-i-MX-8M/m-p/1742476#M214389 <P>Please refer the Machine Learning Guide:&nbsp;<A href="https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf" target="_blank">https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf</A></P> <P>&nbsp;</P> <P>1.We recommend you use tflite model.</P> <P>2.No, we add TFlite delegate support about NPU, but we don't provide any API like CUDA.</P> Thu, 19 Oct 2023 02:24:16 GMT Zhiming_Liu 2023-10-19T02:24:16Z https://community.nxp.com/t5/i-MX-Processors/GPU-Acceleration-and-Parallel-Processing-Capabilities-on-i-MX-8M/m-p/1741783#M214234 <P>Hello,</P><P>I am working on a project that utilizes NXP's i.MX 8M Plus Quad chip. I am exploring the best ways to leverage the GPU of this chip for general-purpose parallel computations. I have a few questions regarding this:</P><OL><LI>What software tools and libraries do you recommend for general-purpose GPU computations on the i.MX 8M Plus Quad?</LI><LI>Is there an opportunity to use a platform or API similar to CUDA? If not, which alternatives would you suggest?</LI><LI>For using the NPU on the chip for artificial intelligence and deep learning applications, which libraries and tools do you recommend?</LI></OL><P>I am open to any suggestions to achieve the best performance for my project. Thanks in advance to everyone who can assist!</P><P><SPAN>Best regards.</SPAN></P> Wed, 18 Oct 2023 05:14:46 GMT https://community.nxp.com/t5/i-MX-Processors/GPU-Acceleration-and-Parallel-Processing-Capabilities-on-i-MX-8M/m-p/1741783#M214234 Tmermer1 2023-10-18T05:14:46Z https://community.nxp.com/t5/i-MX-Processors/GPU-Acceleration-and-Parallel-Processing-Capabilities-on-i-MX-8M/m-p/1742476#M214389 <P>Please refer the Machine Learning Guide:&nbsp;<A href="https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf" target="_blank">https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf</A></P> <P>&nbsp;</P> <P>1.We recommend you use tflite model.</P> <P>2.No, we add TFlite delegate support about NPU, but we don't provide any API like CUDA.</P> Thu, 19 Oct 2023 02:24:16 GMT https://community.nxp.com/t5/i-MX-Processors/GPU-Acceleration-and-Parallel-Processing-Capabilities-on-i-MX-8M/m-p/1742476#M214389 Zhiming_Liu 2023-10-19T02:24:16Z