Hi,
I’m using the STM32N6570-DK (STM32N6).
I’m having an issue with the STM32N6 deployment.
I trained a simple image classification model (bottle vs key), but when I select the ST STM32N6 deployment section, it shows an error and cannot generate the deployment.
My board, ST-Link and STMCubeIDE are all working correctly, so the issue appears to be with theSTEM32N6 deployment.
How can i resolve this error?
Hello @shivi8,
Let me check with our embedded team and come back to you.
Also, could you share your project ID please, this way, I’ll be able to have a look at the configuration of your project and the logs.
Best,
Louis
Hello Louis,
Thank you for getting back to me.
Here’s my project Id : 1073519
Regards,
Shivi
Hi @shivi8
the error is from the ST Edge AI tool when trying to convert the tflite model with the Neural Art compiler from ST.
Converting model…
stedgeai generate --model trained.tflite --target stm32n6 --st-neural-art default@user_neuralart.json -t tflite
ST Edge AI Core v4.0.0-20500 359356bb0
Job scheduled at 29 Jul 2026 11:12:43
INTERNAL ERROR: Exported ONNX could be malformed since ONNX shape inference fails
I’m not sure what can be wrong, first time I see this error.
I’ll try to check if I can get any info.
regards,
fv
