Fails to load!

#1
by cgato - opened

Downloading these ggufs and attempting to load results in.

llama_model_load: error loading model: done_getting_tensors: wrong number of tensors; expected 292, got 291
llama_load_model_from_file: failed to load model
21:33:52-518595 ERROR Failed to load the model.

I also tried converting locally and checked the tensor count and saw 292... but received the same error when loading the model. Not sure what the issue is.

You need a current version of llama.cpp (not older than a few hours) to convert/run llama-3.1 based models. Other than that, there seems nothing wrong with your model - I just tested the Q2_K quant and it worked fine.

mradermacher changed discussion status to closed

Thank you! I'll try it.

Sign up or log in to comment