Nvidia Tesla p40 24GB · Issue #1374 · vllm-project/vllm – GitHub

Hello! Has anyone used GPU p40? I'm interested to know how many tokens it generates per second. Preferably on 7B models.
View full source