When
#1
by MishaGGG - opened
When
Maybe its still quanting, or it ran into a issue, dont have time to check logs rn.
ok
is that model custom parameters or something? seems to be incompatible with llama cpp
warning: token_embd.weight - ncols 584 not divisible by 32 (required for type q8_0) llama_model_quantize: failed to quantize: no tensor type fallback is defined for type q8_0
it s a GPT-like model
And llamacpp has a set of predefined models and sizes, which is out of control for me. The only thing I can do is requeue and hope it works