We've just finished migrating from Hetzner's dedicated servers in Germany to their Cloud US VM's for improved latency. You don't get the same raw performance as their dedicated servers, but they're VMs are still good performers that still ended up being the best value US cloud provider we found [1]
I'll be sure to take a look, but we're really depending on the "high-spec" part as well as the "low RTT" part. IIRC Hetzner's cloud offerings only go up to 48vCPU+192GB memory+1TB disk, all of which are well below our needs.
We're currently using these at OVH: https://www.ovhcloud.com/en-ca/bare-metal/high-grade/hgr-hci... — and we really need the cores, the memory, the bandwidth, and the huge gobs of direct-attached NVMe. (We do highly-concurrent realtime analytics; these machines run DBs that each host thousands of concurrent multi-second OLAP queries against multi-TB datasets, with basically zero temporal locality between queries. It'd actually be a perfect use-case for a huge honking NUMA mainframe with "IO accelerator" cards, but there isn't an efficient market for mainframes—so they're not actually price-optimal here compared to a bunch of replicated DB shards running on commodity hardware.)
[1] https://servicestack.net/blog/finding-best-us-value-cloud-pr...