ai-muninn
Research notes on running LLMs locally: FP8/NVFP4 quantization, DGX Spark, vLLM, and autonomous agents. The real numbers, the traps, and what actually works.
Research notes on running LLMs locally: FP8/NVFP4 quantization, DGX Spark, vLLM, and autonomous agents. The real numbers, the traps, and what actually works.