nih23 comments

Repositories
Issues
Comments

Results 2 comments of


                                            nih23

multi-gpu inference produces broken output

Yes! I just tested different commits to narrow down the issue: Multi-gpu inference has worked fine even on 8 GPUs until (including) 8b428c9bc84be6887d904600d1298b28baffd552. It seems that from 111163e2463171891680feed94371eb9becd9817 something has...

multi-gpu inference produces broken output

Nope, the linux system and llama.cpp are directly accessing the hardware, i.e. no virtualisation is involved.