No ''Available TensorRT Engine Profiles''

[nvidia]

No ''Available TensorRT Engine Profiles''

[RoombaVoid]

I solved the issue by disabling --api. It was enabled by default. I actually dont know what it does exacly but i disabled that and it worked.

Thank you that fixed my Unet not showing up.

[contentis]

I've documented that --api can cause issues when exporting in the readme now. I'll close as this seems to be resolved. Feel free to reopen is you still encounter issues.

[racerx2oo3]

It looks like your engine profiles didn't get written into the model.json file, I suspect that if you open the model.json file it will empty.

You can try deleting all files in the Unet-trt folder including the model.json file and then rebuild your trt engines.

[sprk34]

It looks like your engine profiles didn't get written into the model.json file, I suspect that if you open the model.json file it will empty.

You can try deleting all files in the Unet-trt folder including the model.json file and then rebuild your trt engines.

I tried but this time it gaves me this error

ONNX-Runtime is not installed, so constant folding may be suboptimal or not work at all.
Consider installing ONNX-Runtime: C:\Users\DjSpa\AppData\Roaming\StabilityMatrix\Packages\stable-diffusion-webui\venv\Scripts\python.exe -m pip install onnxruntime
[I] Folding Constants | Pass 1
[!] Module: 'onnxruntime.tools.symbolic_shape_infer' is required but could not be imported.
Note: Error was: No module named 'onnxruntime'
You can set POLYGRAPHY_AUTOINSTALL_DEPS=1 in your environment variables to allow Polygraphy to automatically install missing modules.
[I] Total Nodes | Original: 8992, After Folding: 6216 | 2776 Nodes Folded
[I] Folding Constants | Pass 2
[W] colored module is not installed, will not use colors when logging. To enable colors, please install the colored module: python3 -m pip install colored
[W] Inference failed. You may want to try enabling partitioning to see better results. Note: Error was:
No module named 'onnxruntime'
[!] Module: 'onnxruntime.tools.symbolic_shape_infer' is required but could not be imported.
Note: Error was: No module named 'onnxruntime'
You can set POLYGRAPHY_AUTOINSTALL_DEPS=1 in your environment variables to allow Polygraphy to automatically install missing modules.
[I] Total Nodes | Original: 6216, After Folding: 6152 | 64 Nodes Folded
[I] Folding Constants | Pass 3
[W] colored module is not installed, will not use colors when logging. To enable colors, please install the colored module: python3 -m pip install colored
[W] Inference failed. You may want to try enabling partitioning to see better results. Note: Error was:
No module named 'onnxruntime'
[!] Module: 'onnxruntime.tools.symbolic_shape_infer' is required but could not be imported.
Note: Error was: No module named 'onnxruntime'
You can set POLYGRAPHY_AUTOINSTALL_DEPS=1 in your environment variables to allow Polygraphy to automatically install missing modules.
[I] Total Nodes | Original: 6152, After Folding: 6152 | 0 Nodes Folded

I'll try to delete all the files that comes with tensort-rt in''venv'' and will reinstall it. Thanx for reply.

[racerx2oo3]

Unless the engine generation actually failed, those errors are more likely just warnings and things should still work as expected.

[sprk34]

Unless the engine generation actually failed, those errors are more likely just warnings and things should still work as expected.

Didn't know that thank you 👍 but still no luck, You are right about ''model json'' file. It is still empty.I have no clue what the problem and will never understand probably:) Thank you anyway

--I forgot the say this. When i trying to export an engine it gaves this error--
''[INFO]: TensorRT engine found. Skipping build. You can enable Force Export in the Advanced Settings to force a rebuild if needed.''
I am sending this maybe its bug or something. It says found but no engines under the profile sections.

[sprk34]

I solved the issue by disabling --api. It was enabled by default. I actually dont know what it does exacly but i disabled that and it worked.

[contentis]

''[INFO]: TensorRT engine found. Skipping build. You can enable Force Export in the Advanced Settings to force a rebuild if needed.''

To avoid this, you'll need to delete the engine from Unet-trt.

I have not yet reproduced this issue. I'll work on this as soon as possible, but due to the amount of issues and limited time this might need some time to be fixed.