qi-train

Convert, fine-tune or train weights for qi — chiefly converting IBM's activated LoRA safetensors to GGUF so the answerability gate costs milliseconds instead of seconds, and the pre-Granite research lane under train/. Use when asked to convert an adapter, produce activated adapters for a new model size, fine-tune anything, run a training job, or touch train/, and read it before running any GPU work for this project.

License

AGPL-3.0-or-later

Compatibility

Conversion needs Python with transformers and a llama.cpp checkout. Training needs a real GPU — never a Mac. Modal or a CUDA box.