Closed
Description
Expected Behavior
server.cpp should recognise parameters -tb / --threads-batch (as stated in the readme).
Please provide a detailed written description of what llama.cpp
did, instead.
server.cpp doesn't recognise the -tb / --threads-batch parameter.
I checked the code, this options seems indeed missing.
PS: I can attempt adding it, if you agree... it would be a good task to get started on the code.