Unsloth releases Dynamic v3.0 quants for Qwen3.8-27B

Unsloth released Dynamic v3.0 quantized versions of Qwen3.8-27B, replacing an earlier shared preview. The company said the GGUF files work with most inference engines, including llama.cpp and Unsloth Desktop, and use post-training quantization rather than quantization-aware training. Unsloth said its new calibration dataset, layer selection and quantization methods preserve more model quality at a given disk size. It also said smaller quantizations remove the MTP module to save about 500MB of disk space.