irresistiblegrace97/TinyLlama-1.1B-Chat-v1.0-torchao-int4_weight_only-gs_4096. It has been quantized using int4_weight_only quantization with torchao.1from transformers import AutoModel
2
3model = AutoModel.from_pretrained("irresistiblegrace97/TinyLlama-1.1B-Chat-v1.0-torchao-int4_weight_only-gs_4096")