D
Nov 5, 2026DIY Self-Hosting 35B+ Reasoning LLMs with NVFP4 Quantization and vLLM
A practical hands-on blueprint for deploying 35B and 70B reasoning models locally using NVFP4 micro-scaling floating-point quantization and vLLM on personal workstations.
Nov 5, 20263 min read0 reactions1 comments