Private-BitSix-Mistral-Small-3.1-24B-Instruct-2503 is a model optimized to run on local 4090 GPUs through 6-bit quantization, based on Mistral-Small-3.1-24B-Instruct-2503