Error with quantization and lora

[imikod]

When I use a model with lora it works as expected but when I pass --type q8_0 it does not work at all.
The model works without lora and with quantization.
I tested different loras and models with the same result.

[FSSRepo]

The results of Loras with quantized models are very poor anyway. I do not recommend it.

[imikod]

I believe you. In fact I read the warning and wanted to see how bad it was.

[leejet]

This PR #969 should fix the issue.