Hacker News new | ask | show | jobs
by orbanlevi 34 days ago
I have 1 DGX Spark and running models with vLLM to, out of curiosity why not using Llama.cpp / TensorRT-LLM or any other alternatives?