[warning] loading model fails!

[Godlovecui]

Hello，
The quantized file suffix is ".safetensor", however, when I execute ./bin/llama_example, it output the below warning info，

where can I get these ".bin" files?
Thank you!

[lswzjuer]

Hello, where did you get the W8A16 model weights? We open source the simulated quantized ckpts, which is only used for accuracy verification. If actual inference is performed, packing preprocessing is required.