David Castillo
Running vLLM on Consumer Hardware: The VRAM Math, Memory Management, and Why It Beats llama.cpp
How I run vLLM on the same consumer rig as my llama.cpp setup: the real VRAM budget (weights, MTP, KV cache, PyTorch overhead), why the repo...
Running Local LLMs on Consumer Hardware: A Practical Guide
How I deployed and optimized local LLMs on a 5-year-old consumer PC using llama.cpp, from hardware bottlenecks to advanced inference techniq...