import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

# 加载模型和分词器
model_name = "meta-llama/Llama-2-7b-hf"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto")

# 定义生成函数
def generate_text(prompt, max_length=256, temperature=0.7):
    inputs = tokenizer(prompt, return_tensors="pt", truncation=True)
    outputs = model.generate(
        **inputs,
        max_length=max_length,
        temperature=temperature,
        num_return_sequences=1,
    )
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# 创建 Gradio 界面
interface = gr.Interface(
    fn=generate_text,
    inputs=[
        gr.Textbox(label="输入 Prompt", placeholder="请输入您的 Prompt..."),
        gr.Slider(50, 512, step=10, value=256, label="最大生成长度"),
        gr.Slider(0.1, 1.0, step=0.1, value=0.7, label="生成温度"),
    ],
    outputs="text",
    title="Llama-2 Text Generator",
    description="使用 meta-llama/Llama-2-7b-hf 进行文本生成。",
)

# 启动 Gradio 应用
if __name__ == "__main__":
    interface.launch()