[halJordan]

Quantization is an extraordinarily trivial process. Especially if you're doing it with llama.cpp (which unsloth obviously does).

Qwen did release an fp8 version, which is a quantized version.