Upload merged Qwen2.5-7B-Instruct model with LoRA adapter (BF16) 5defbe6 verified mssongit commited on Oct 16