How to run it on a mobile device?
#1
by
KoiSikhaDo
- opened
A lot of the non-allenai blogs online say that the model is small enough to run on a mobile device. Just wanted to know if this can be done.
Can the model be quantised using bitsandbytes to make it smaller to run on a mobile device?
KoiSikhaDo
changed discussion status to
closed
KoiSikhaDo
changed discussion status to
open
Yes actually @soldni ran https://huggingface.co./allenai/OLMoE-1B-7B-0924-GGUF on a mobile device - not sure if there is a public guide about it somewhere?
We'll also need to merge MolmoE into llama.cpp first to make it work like the above one (i.e. a PR like https://github.com/ggerganov/llama.cpp/pull/9462)
stay tuned, we are trying to get this one running!