The biggest perf improvement in Valkey, will be RDMA for the version 8 release. See https://github.com/valkey-io/valkey/pull/477#issuecomment-21...
One of the strategies that is being used in the multi-threading is using CPU memory prefetching to pull memory closing to the CPU so that we aren't stalling on fetching data from main memory while executing commands. We still want to try to apply these techniques without multi-threading to improve the efficiency of single or double core installations.
I would expect vertical scaling to have better performance for most use cases as you can get cheap servers with very large amounts of RAM and high number of CPU cores.