nm-testing/Llama-4-Scout-17B-16E-Instruct-for-quant

This is unquantized Llama-4-Scout model which has been converted into a format that can be used for quantization. The main diff is that it unrolls experts into separate modules so they can be picked up independently by quantizers. It has been used to produce RedHatAI quantized models: https://huggingface.co/RedHatAI/Llama-4-Scout-17B-16E-Instruct-FP8-dynamic, https://huggingface.co/RedHatAI/Llama-4-Scout-17B-16E-Instruct-quantized.w4a16