dohnuts.cpp

A native C++ CPU-only port of the Dohnuts System One model, built on llama.cpp, that runs a merged Qwen3.5-0.8B-plus-LoRA model and returns decision probabilities from a single forward pass with no GPU and no text generation involved.

unverified — sourced from awesomejev.com; description independently written by AionEdge from the repo’s own GitHub metadata and README, not GitHub-API relevance-checked