GGUF

A 2016 Intel Xeon server with 128 GB DDR3 RAM and no GPU runs a 26B Mixture-of-Experts model using CPU-optimized inference and a long, flag-heavy tuning process, illustrating memory-bandwidth limits and the claimed viability of open-weight AI on commodity hardware.

A Metal-only local inference engine (ds4.c) for DeepSeek V4 Flash with 1M-token context, 2-bit quantization, and disk-backed KV cache, offering OpenAI/Anthropic-compatible local APIs but with alpha-quality code and very high RAM requirements on macOS.