Output fails with " Error message from AI engine: model failed."

[Skwerl (Josh)]

1. The bug / behavior you have encountered
   Twice, I’ve started a long encode only to have it error out for some reason hours later. This might have something to do with a device context loss due to me using remote desktop to access the machine occasionally. The error message is too generic for me to say whether this is a duplicate of another report: Keep getting Error model failed

2. Your system profile
   Intel i7-13700K, 64 GB RAM, NVidia RTX 3080 Windows 11 Pro

3. Your log files (Help > Logging > Get Logs for Support)
   logsForSupport.zip

[kyle.topazlabs]

The logs are pointing to an ONNX error when running the export, but it could also be an issue with the audio files on the export.

Swap the audio mode to convert and then open the Preferences Panel and change the following settings.

File>Preferences>Processing

1. Ensure that your GPU is selected for the AI Processor
2. Set the max memory to 80%
3. Set max processes to 1
4. Save the settings and close the panel.
5. Re-run your clip through the process that you were trying.

If you run into an issue again, please send me the updated logs for further review. help@topazlabs.com

[Skwerl (Josh)]

Thanks, Kyle. I appreciate the quick response, This is a UHD movie, so I don’t really relish the thought of converting the master audio. The first time this error occurred, it was almost at the end of the processing (for two days!). I would imagine if the audio were a problem, the crash would be more deterministic. I’ll just disable audio altogether and then remultiplex it back in later.
I’ve decreased max memory to 80% and will try again (the other items are already configured as you suggested). Hoping I have some luck!