Show HN: Proxima serves 4x more requests with no hardware change on vLLM
Proxima lets existing GPUs serve 4x more concurrent requests - Tenosra/Proxima
Summary not yet generated for this article.
Original Text
hey everyone, i decided to make a vLLM plugin that implements the Star-KV paper. the results are quiet promising with a decode kernel thats faster than FA2 in higher batch sizes. would love any thoughts and recommendations
Comments (1 of ) ⌛