Not sure if this is of any value to you, but Ryzen 7 generates 2 tokens per second for the 7B-Instruct model. The model itself is very unimpressive and I see n

Not sure if this is of any value to you, but Ryzen 7 generates 2 tokens per second for the 7B-Instruct model. The model itself is very unimpressive and I see no reason to play with it over the worst alternative from Hugging Face. I can only imagine this was released for some bizarre compliance reasons.