from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "killer66678/qwen2.5_32b_lora"
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

tokenizer = AutoTokenizer.from_pretrained(model_id)

prompt = "Hello"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=64)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Notes

Training data and fine-tuning hyperparameters are not disclosed.

本研究的实验与计算工作依托于华为云昇腾AI云服务平台完成，特此对其提供的稳定算力支持表示感谢。