The benchmarks behind the launch, raw data included, plus the one we didn't win

Everything on the launch page comes from a measurement you can rerun yourself. The harness, the raw memtier output and every flag we set are in the public repo: https://github.com/hotkv-ltd/hotkv

I also recorded two sessions on a single 160 vCPU Azure box, same memtier line, one server after the other, with the load generator on the same machine (so the absolute numbers are lower than the published run):

HotKV vs Redis 8.10.2: https://youtu.be/dLrqD42-xNI

HotKV vs Valkey 9.1.2: https://youtu.be/F8Z_I9zP7s4

The published benchmark used separate load generators and three runs per command. There HotKV had the highest throughput on 11 of 12 commands against Redis, Valkey and Dragonfly. The 12th was PING, which was basically a tie. Dragonfly came out 0.6% ahead with the load generators maxed out, and that's in the report as well.

If you run the harness on your own hardware and see something different, post it here. Please include hardware, versions, the exact commands and the raw output so people can check it. Happy to answer questions about the method or the engine.