Text Generation
Transformers
Safetensors
English
llama
text-generation-inference
4-bit precision

ggml version?

#3
by cooki3monster - opened

Hi,

Do you have or plan to have the ggml model that runs on llamacpp CPU.

Thanks

https://huggingface.co/TheBloke/wizard-mega-13B-GGML

but you probably want https://huggingface.co/TheBloke/Manticore-13B-GGML ? I think it’s the same one, but a later epoch. They changed the name.

Awesome already using multicore-13B. Thanks for confirming it's the equivalent.

Yes Manticore replaces Wizard Mega. No reason to use Wizard Mega now that Manticore is out, I believe.

Sign up or log in to comment