Anonymous
8/20/2025, 9:30:33 AM
No.106321837
>>106321808
Yes and it goes in GPU by default when you pass it as a parameter, unless you use the "--no-mmproj-offload" flag which would send it to CPU instead - in llama.cpp "offload" means offloading from CPU to GPU, not the other way around like it's usually used
Yes and it goes in GPU by default when you pass it as a parameter, unless you use the "--no-mmproj-offload" flag which would send it to CPU instead - in llama.cpp "offload" means offloading from CPU to GPU, not the other way around like it's usually used