llama.cpp support

#1
by djuna - opened

Will you add this model architecture into llama.cpp?

See ggml-org/llama.cpp#15510.

I see this still has not been merged. Is this adding full support and merge ready? What's taking so long?

I see this still has not been merged. Is this adding full support and merge ready? What's taking so long?

It was a draft for a while pending a necessary GGML feature, it is now just pending review. In the mean time, if you can, please build it yourself and test, it is fully functional.

Looks like it's been merged now FYI

Nice. Thanks lcpp team.

djuna changed discussion status to closed

Sign up or log in to comment