chickadee » llama » make-am-inference-backend

make-am-inference-backend #!key threadsprocedure

Returns the crunch inference backend, for run-with-backend. threads, when given, is the number of threads used by the crunch kernels.